{smcl}
{com}{sf}{ul off}{txt}{.-}
      name:  {res}<unnamed>
       {txt}log:  {res}C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\OUTPUT.Organizational Memory.MANUSCRIPT.04-30-2025.smcl
  {txt}log type:  {res}smcl
 {txt}opened on:  {res}30 Apr 2025, 19:25:16
{txt}
{com}. 
. *******************************************************************************************************************************************************************************************************************************
. use "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\2024June Data\FINAL DATABASE\Organizational Memory.FINAL DATABASE.04-29-2025.dta", replace
{txt}
{com}. 
. 
. *
. *
. *
. 
. egen a_id = group(bureau_opm)
{txt}
{com}. 
. * COMPUTE: Compute the FE estimator as a pooled OLS estimator using the Mundlak device (Papke and Wooldridge 2008)
. egen turnoverb = mean(turnover), by(a_id)
{txt}
{com}. egen itsepabove15_los15pctb = mean(itsepabove15_los15pct), by(a_id)
{txt}
{com}. 
. egen turnover_gs13b = mean(turnover_gs13), by(a_id)
{txt}(4 missing values generated)

{com}. egen gs13above15_los15pctb = mean(gs13above15_los15pct), by(a_id)
{txt}(3 missing values generated)

{com}. 
. egen voluntaryb = mean(voluntary), by(a_id)
{txt}
{com}. egen volabove15_los15pctb = mean(volabove15_los15pct), by(a_id)
{txt}
{com}. 
. egen involuntaryb = mean(involuntary), by(a_id)
{txt}
{com}. egen involabove15_los15pctb = mean(involabove15_los15pct), by(a_id)
{txt}
{com}. 
. egen voluntary_gs13b = mean(voluntary_gs13), by(a_id)
{txt}(4 missing values generated)

{com}. egen volgs13_los15pctb = mean(volgs13_los15pct), by(a_id)
{txt}(3 missing values generated)

{com}. 
. egen involuntary_gs13b = mean(involuntary_gs13), by(a_id)
{txt}(4 missing values generated)

{com}. egen involgs13_los15pctb = mean(involgs13_los15pct), by(a_id)
{txt}(3 missing values generated)

{com}. 
. 
. egen activityno_permil_pctb = mean(activityno_permil_pct), by(a_id)
{txt}
{com}. egen cio_tenureb = mean(cio_tenure), by(a_id)
{txt}
{com}. egen lnitempb = mean(lnitemp), by(a_id)
{txt}
{com}. egen lnprojsizeb = mean(lnprojsize), by(a_id)
{txt}
{com}. egen projects_nob = mean(projects_no), by(a_id)
{txt}
{com}. egen accrateb = mean(accrate), by(a_id)
{txt}
{com}. egen newprojects_pctb = mean(newprojects_pct), by(a_id)
{txt}
{com}. egen dmeratio_totalspendingb = mean(dmeratio_totalspending), by(a_id)
{txt}(7 missing values generated)

{com}. 
. 
. save "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\DATA\Organizational Memory.STATISTICAL DATABASE.04-30-2025.dta", replace
{txt}{p 0 4 2}
file {bf}
C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\DATA\Organizational Memory.STATISTICAL DATABASE.04-30-2025.dta{rm}
saved
{p_end}

{com}. 
. *******************************************************************************************************************************************************************************************************************************
. 
. quietly eststo: xtgee delay_pct c.turnover itsepabove15_los15pct i.cio5 activityno_permil_pct cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnoverb itsepabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   dmeratio_totalspending dmeratio_totalspendingb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}
{com}. 
. /* Set global */
. sum turnover if e(sample), detail

                          {txt}turnover2
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res} .0055744              0       {txt}Obs         {res}        540
{txt}25%    {res} .0188012              0       {txt}Sum of wgt. {res}        540

{txt}50%    {res} .0357143                      {txt}Mean          {res} .0451436
                        {txt}Largest       Std. dev.     {res} .0418133
{txt}75%    {res} .0599685            .25
{txt}90%    {res} .0910686       .2592593       {txt}Variance      {res} .0017484
{txt}95%    {res} .1181108       .2901554       {txt}Skewness      {res} 2.415703
{txt}99%    {res} .2060086       .3389831       {txt}Kurtosis      {res} 12.78484
{txt}
{com}. global turnover_p10 = round(r(p10),0.01)
{txt}
{com}. global turnover_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum turnover_gs13 if e(sample), detail

                        {txt}turnover_gs13
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        522
{txt}25%    {res} .0071556              0       {txt}Sum of wgt. {res}        522

{txt}50%    {res} .0188934                      {txt}Mean          {res} .0296624
                        {txt}Largest       Std. dev.     {res} .0348858
{txt}75%    {res} .0403226             .2
{txt}90%    {res} .0732601       .2222222       {txt}Variance      {res}  .001217
{txt}95%    {res} .0909091            .25       {txt}Skewness      {res} 2.571793
{txt}99%    {res} .1716738            .25       {txt}Kurtosis      {res} 12.67952
{txt}
{com}. global turnovergs13_p10 = round(r(p10),0.01)
{txt}
{com}. global turnovergs13_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum voluntary if e(sample), detail

                          {txt}voluntary
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res} .0048931              0       {txt}Obs         {res}        540
{txt}25%    {res} .0166252              0       {txt}Sum of wgt. {res}        540

{txt}50%    {res} .0326397                      {txt}Mean          {res} .0416391
                        {txt}Largest       Std. dev.     {res} .0372932
{txt}75%    {res} .0570524       .2060086
{txt}90%    {res} .0833333            .25       {txt}Variance      {res} .0013908
{txt}95%    {res} .1111063            .25       {txt}Skewness      {res} 2.040546
{txt}99%    {res} .1764706       .2592593       {txt}Kurtosis      {res} 9.771108
{txt}
{com}. global voluntary_p10 = round(r(p10),0.01)
{txt}
{com}. global voluntary_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum involuntary if e(sample), detail

                         {txt}involuntary
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        540
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        540

{txt}50%    {res}        0                      {txt}Mean          {res} .0034893
                        {txt}Largest       Std. dev.     {res} .0123897
{txt}75%    {res} .0016282       .0526316
{txt}90%    {res} .0082647       .0769231       {txt}Variance      {res} .0001535
{txt}95%    {res} .0204167       .1554404       {txt}Skewness      {res} 8.547913
{txt}99%    {res} .0387724       .1638418       {txt}Kurtosis      {res} 98.56224
{txt}
{com}. global involuntary_p10 = round(r(p10),0.01)
{txt}
{com}. global involuntary_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum voluntary_gs13 if e(sample), detail

                       {txt}voluntary_gs13
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        522
{txt}25%    {res} .0066335              0       {txt}Sum of wgt. {res}        522

{txt}50%    {res} .0175696                      {txt}Mean          {res} .0291636
                        {txt}Largest       Std. dev.     {res} .0347796
{txt}75%    {res} .0396825             .2
{txt}90%    {res} .0724638       .2222222       {txt}Variance      {res} .0012096
{txt}95%    {res} .0909091            .25       {txt}Skewness      {res} 2.587143
{txt}99%    {res} .1716738            .25       {txt}Kurtosis      {res} 12.85123
{txt}
{com}. global voluntary_gs13_p10 = round(r(p10),0.01)
{txt}
{com}. global voluntary_gs13_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum involuntary_gs13 if e(sample), detail

                      {txt}involuntary_gs13
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        522
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        522

{txt}50%    {res}        0                      {txt}Mean          {res} .0004831
                        {txt}Largest       Std. dev.     {res} .0030796
{txt}75%    {res}        0       .0148148
{txt}90%    {res} .0005167       .0208333       {txt}Variance      {res} 9.48e-06
{txt}95%    {res}  .001845       .0333333       {txt}Skewness      {res} 12.56043
{txt}99%    {res} .0075188       .0526316       {txt}Kurtosis      {res} 187.9219
{txt}
{com}. global involuntary_gs13_p10 = round(r(p10),0.01)
{txt}
{com}. global involuntary_gs13_p90 = round(r(p90),0.01)
{txt}
{com}. *
. *
. *
. sum itsepabove15_los15pct if e(sample), detail

                    {txt}itsepabove15_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res} .0208696              0       {txt}Obs         {res}        540
{txt}25%    {res} .0555556              0       {txt}Sum of wgt. {res}        540

{txt}50%    {res} .0807887                      {txt}Mean          {res} .0892419
                        {txt}Largest       Std. dev.     {res}  .073472
{txt}75%    {res} .1102423             .5
{txt}90%    {res} .1428571             .5       {txt}Variance      {res} .0053981
{txt}95%    {res} .1948276       .5833333       {txt}Skewness      {res} 5.080016
{txt}99%    {res} .2857143              1       {txt}Kurtosis      {res} 52.91218
{txt}
{com}. global los_p10 = round(r(p10),0.01)
{txt}
{com}. global los_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum volabove15_los15pct if voluntary>0 & e(sample), detail

                     {txt}volabove15_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        487
{txt}25%    {res} .0012484              0       {txt}Sum of wgt. {res}        487

{txt}50%    {res} .0121528                      {txt}Mean          {res} .0247642
                        {txt}Largest       Std. dev.     {res} .0620929
{txt}75%    {res} .0260223            .25
{txt}90%    {res} .0520833             .5       {txt}Variance      {res} .0038555
{txt}95%    {res} .0833333             .5       {txt}Skewness      {res} 10.19587
{txt}99%    {res}      .25              1       {txt}Kurtosis      {res} 140.6657
{txt}
{com}. global vollos_p10 = round(r(p10),0.01)
{txt}
{com}. global vollos_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum involabove15_los15pct if involuntary>0 & e(sample), detail

                    {txt}involabove15_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        186
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        186

{txt}50%    {res}        0                      {txt}Mean          {res} .0022848
                        {txt}Largest       Std. dev.     {res} .0053594
{txt}75%    {res} .0011971       .0229401
{txt}90%    {res} .0087719       .0285714       {txt}Variance      {res} .0000287
{txt}95%    {res}  .012987       .0294118       {txt}Skewness      {res} 3.181416
{txt}99%    {res} .0294118         .03125       {txt}Kurtosis      {res} 14.26383
{txt}
{com}. global invollos_p10 = round(r(p10),0.01)
{txt}
{com}. global invollos_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum gs13above15_los15pct if turnover_gs13>0 & e(sample), detail

                    {txt}gs13above15_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        453
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        453

{txt}50%    {res} .0082305                      {txt}Mean          {res}  .021638
                        {txt}Largest       Std. dev.     {res} .0635654
{txt}75%    {res} .0196078            .25
{txt}90%    {res} .0481928             .5       {txt}Variance      {res} .0040406
{txt}95%    {res} .0819672             .5       {txt}Skewness      {res} 10.31437
{txt}99%    {res}      .25              1       {txt}Kurtosis      {res} 139.6683
{txt}
{com}. global gs13_los_p10 = round(r(p10),0.01)
{txt}
{com}. global gs13_los_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum volgs13_los15pct if voluntary_gs13>0 & e(sample), detail

                      {txt}volgs13_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        450
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        450

{txt}50%    {res} .0076923                      {txt}Mean          {res} .0213994
                        {txt}Largest       Std. dev.     {res} .0638118
{txt}75%    {res} .0192308            .25
{txt}90%    {res} .0482899             .5       {txt}Variance      {res} .0040719
{txt}95%    {res} .0819672             .5       {txt}Skewness      {res} 10.27428
{txt}99%    {res}      .25              1       {txt}Kurtosis      {res} 138.5985
{txt}
{com}. global volgs13_los_p10 = round(r(p10),0.01)
{txt}
{com}. global volgs13_los_p90 = round(r(p90),0.01)
{txt}
{com}. 
. sum involgs13_los15pct if involuntary_gs13>0 & e(sample), detail

                     {txt}involgs13_los15pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}         79
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}         79

{txt}50%    {res}  .000399                      {txt}Mean          {res} .0021808
                        {txt}Largest       Std. dev.     {res} .0044706
{txt}75%    {res} .0020067        .012987
{txt}90%    {res} .0078431       .0132275       {txt}Variance      {res}   .00002
{txt}95%    {res}  .012987       .0136986       {txt}Skewness      {res} 3.482428
{txt}99%    {res} .0285714       .0285714       {txt}Kurtosis      {res} 17.82256
{txt}
{com}. global involgs13_los_p10 = round(r(p10),0.01)
{txt}
{com}. global involgs13_los_p90 = round(r(p90),0.01)
{txt}
{com}. *
. *
. *
. sum activityno_permil_pct if e(sample), detail

                    {txt}activityno_permil_pct
{hline 61}
      Percentiles      Smallest
 1%    {res}        0              0
{txt} 5%    {res}        0              0
{txt}10%    {res}        0              0       {txt}Obs         {res}        540
{txt}25%    {res}        0              0       {txt}Sum of wgt. {res}        540

{txt}50%    {res} .2222222                      {txt}Mean          {res} .2909892
                        {txt}Largest       Std. dev.     {res} .2820347
{txt}75%    {res}  .494898              1
{txt}90%    {res} .7071429              1       {txt}Variance      {res} .0795436
{txt}95%    {res} .9166667              1       {txt}Skewness      {res} .8906631
{txt}99%    {res}        1              1       {txt}Kurtosis      {res} 2.992449
{txt}
{com}. global stdz_p10 = round(r(p10),0.01)
{txt}
{com}. global stdz_p90 = round(r(p90),0.01)
{txt}
{com}. 
. ***********************************************
. * DESCRIPTIVES
. xtsum turnover turnover_gs13 voluntary involuntary voluntary_gs13 involuntary_gs13  itsepabove15_los15pct volabove15_los15pct involabove15_los15pct gs13above15_los15pct volgs13_los15pct involgs13_los15pct cio5 activityno_permil_pct  lnitemp lnprojsize projects_no accrate  newprojects_pct cio_tenure dmeratio_totalspending if e(sample)

{txt}Variable         {c |}      Mean   Std. dev.       Min        Max {c |}    Observations
{hline 17}{c +}{hline 44}{c +}{hline 16}
turnover{col 10}overall {c |} {res} .0451436   .0418133          0   .3389831{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0322721          0   .1768531{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0287364  -.0798564   .2072736{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
turno~13{col 10}overall {c |} {res} .0296624   .0348858          0        .25{txt} {c |}{col 69}N =     522
{col 10}between {c |}{col 31}{res} .0386651          0        .25{txt} {c |}{col 69}n =      92
{col 10}within  {c |}{col 31}{res}  .020171  -.0597315   .1542609{txt} {c |} T-bar = 5.67391
{col 18}{c |}{col 63}{c |}
volunt~y{col 10}overall {c |} {res} .0416391   .0372932          0   .2592593{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0290463          0   .1608796{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0267009  -.0833609   .1701441{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
involu~y{col 10}overall {c |} {res} .0034893   .0123897          0   .1638418{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0081828          0   .0668323{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0087969  -.0507644   .1004988{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
volun~13{col 10}overall {c |} {res} .0291636   .0347796          0        .25{txt} {c |}{col 69}N =     522
{col 10}between {c |}{col 31}{res} .0386334          0        .25{txt} {c |}{col 69}n =      92
{col 10}within  {c |}{col 31}{res} .0201878  -.0602303   .1549142{txt} {c |} T-bar = 5.67391
{col 18}{c |}{col 63}{c |}
invol~13{col 10}overall {c |} {res} .0004831   .0030796          0   .0526316{txt} {c |}{col 69}N =     522
{col 10}between {c |}{col 31}{res} .0011805          0   .0075188{txt} {c |}{col 69}n =      92
{col 10}within  {c |}{col 31}{res}  .002826  -.0070357   .0455958{txt} {c |} T-bar = 5.67391
{col 18}{c |}{col 63}{c |}
itsepa~t{col 10}overall {c |} {res} .0892419    .073472          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0459515          0       .375{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0625012  -.2857581   .7142419{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
volabo~t{col 10}overall {c |} {res} .0223336   .0594205          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0468342          0       .375{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0462242  -.3526664   .6473336{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
invola~t{col 10}overall {c |} {res}  .000787   .0033226          0     .03125{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .0020532          0    .015972{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0025846  -.0050954   .0275727{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
gs13ab~t{col 10}overall {c |} {res} .0187419   .0596079          0          1{txt} {c |}{col 69}N =     523
{col 10}between {c |}{col 31}{res} .0646887          0         .5{txt} {c |}{col 69}n =      93
{col 10}within  {c |}{col 31}{res} .0432589  -.3562581   .6437419{txt} {c |} T-bar = 5.62366
{col 18}{c |}{col 63}{c |}
volgs1~t{col 10}overall {c |} {res} .0184125   .0596456          0          1{txt} {c |}{col 69}N =     523
{col 10}between {c |}{col 31}{res} .0647587          0         .5{txt} {c |}{col 69}n =      93
{col 10}within  {c |}{col 31}{res} .0432469  -.3565875   .6434125{txt} {c |} T-bar = 5.62366
{col 18}{c |}{col 63}{c |}
involg~t{col 10}overall {c |} {res} .0003294   .0018967          0   .0285714{txt} {c |}{col 69}N =     523
{col 10}between {c |}{col 31}{res} .0007489          0   .0040816{txt} {c |}{col 69}n =      93
{col 10}within  {c |}{col 31}{res} .0017316  -.0037522   .0248192{txt} {c |} T-bar = 5.62366
{col 18}{c |}{col 63}{c |}
cio5{col 10}overall {c |} {res} .1833333   .3872983          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .2534635          0          1{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .3088285       -.65   1.040476{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
activi~t{col 10}overall {c |} {res} .2909892   .2820347          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .2122087          0   .7708333{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .1991229  -.4590108   1.114519{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
lnitemp{col 10}overall {c |} {res} 5.438428   1.715548   1.386294   9.434044{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} 1.769255   1.386294   9.371997{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .1990504   3.513927   6.188075{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
lnproj~e{col 10}overall {c |} {res} 2.397292   2.202821  -6.907755   8.961861{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} 1.801252  -1.466238   6.017139{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} 1.290115  -7.266694   7.887873{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
projec~o{col 10}overall {c |} {res} 19.82778   28.34157          1        230{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} 23.12078          1   145.5714{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} 14.06991  -109.7437   104.2563{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
accrate{col 10}overall {c |} {res} .0853265   .0713482          0   .3954802{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res}  .049939          0    .250093{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0526348  -.1079508   .3580538{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
newpro~t{col 10}overall {c |} {res} .4823864   .2758069          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res}  .196774          0          1{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .2183685  -.2676136   1.251651{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
cio_te~e{col 10}overall {c |} {res} 2.366114    1.50598   .4958904   10.92603{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .9891436   1.311546   8.523288{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} 1.163516  -.8877676   6.372506{txt} {c |} T-bar = 5.74468
{col 18}{c |}{col 63}{c |}
dmerat~g{col 10}overall {c |} {res} .2150978   .1488404          0          1{txt} {c |}{col 69}N =     540
{col 10}between {c |}{col 31}{res} .1275836          0   .7818091{txt} {c |}{col 69}n =      94
{col 10}within  {c |}{col 31}{res} .0927402  -.4394751   .6833144{txt} {c |} T-bar = 5.74468

{com}. ***********************************************
. 
. sum delay_pct if e(sample)

{txt}    Variable {c |}        Obs        Mean    Std. dev.       Min        Max
{hline 13}{c +}{hline 57}
{space 3}delay_pct {c |}{res}        540    .2940294    .2908181          0          1
{txt}
{com}. di r(sd)
{res}.29081809
{txt}
{com}. 
. discard
{txt}
{com}. 
. *********************************************************************************************************************************************************************************************************************************
. 
. 
. 
. 
. **# [MODELS 1 - 4] [FIGURES 3A-6A] AGGREGATE TURNOVER MODELS
. 
. 
. 
. * [M1, FIGURE 3A] H1: Turnover Baseline Effects
. eststo: xtgee delay_pct c.turnover itsepabove15_los15pct i.cio5 activityno_permil_pct cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnoverb itsepabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.57603939
{txt}Iteration 2:  Tolerance = {res}.05001729
{txt}Iteration 3:  Tolerance = {res}.02139581
{txt}Iteration 4:  Tolerance = {res}.00472287
{txt}Iteration 5:  Tolerance = {res}.00208842
{txt}Iteration 6:  Tolerance = {res}.00093714
{txt}Iteration 7:  Tolerance = {res}.00045027
{txt}Iteration 8:  Tolerance = {res}.00023664
{txt}Iteration 9:  Tolerance = {res}.00011924
{txt}Iteration 10: Tolerance = {res}.00005823
{txt}Iteration 11: Tolerance = {res}.00002773
{txt}Iteration 12: Tolerance = {res}.00001293
{txt}Iteration 13: Tolerance = {res}5.915e-06
{txt}Iteration 14: Tolerance = {res}2.660e-06
{txt}Iteration 15: Tolerance = {res}1.177e-06
{txt}Iteration 16: Tolerance = {res}5.123e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:355.46}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 15}turnover {c |}{col 25}{res}{space 2} 1.913994{col 37}{space 2} 1.121197{col 48}{space 1}    1.71{col 57}{space 3}0.088{col 65}{space 4}-.2835105{col 78}{space 3} 4.111499
{txt}{space 2}itsepabove15_los15pct {c |}{col 25}{res}{space 2}-.4103782{col 37}{space 2} .5783393{col 48}{space 1}   -0.71{col 57}{space 3}0.478{col 65}{space 4}-1.543902{col 78}{space 3} .7231459
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.0462568{col 37}{space 2} .0951321{col 48}{space 1}   -0.49{col 57}{space 3}0.627{col 65}{space 4}-.2327124{col 78}{space 3} .1401988
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2317026{col 37}{space 2} .2242868{col 48}{space 1}    1.03{col 57}{space 3}0.302{col 65}{space 4}-.2078914{col 78}{space 3} .6712967
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0115921{col 37}{space 2} .0269276{col 48}{space 1}   -0.43{col 57}{space 3}0.667{col 65}{space 4}-.0643693{col 78}{space 3} .0411851
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.3902337{col 37}{space 2} .3071107{col 48}{space 1}   -1.27{col 57}{space 3}0.204{col 65}{space 4}-.9921596{col 78}{space 3} .2116922
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0362507{col 37}{space 2}  .029313{col 48}{space 1}    1.24{col 57}{space 3}0.216{col 65}{space 4}-.0212017{col 78}{space 3} .0937031
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0013142{col 37}{space 2} .0016647{col 48}{space 1}    0.79{col 57}{space 3}0.430{col 65}{space 4}-.0019486{col 78}{space 3} .0045771
{txt}{space 16}accrate {c |}{col 25}{res}{space 2}  .058734{col 37}{space 2} .5507784{col 48}{space 1}    0.11{col 57}{space 3}0.915{col 65}{space 4}-1.020772{col 78}{space 3}  1.13824
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8308698{col 37}{space 2} .1705064{col 48}{space 1}   -4.87{col 57}{space 3}0.000{col 65}{space 4}-1.165056{col 78}{space 3}-.4966833
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.252356{col 37}{space 2} .1003075{col 48}{space 1}  -12.49{col 57}{space 3}0.000{col 65}{space 4}-1.448955{col 78}{space 3}-1.055757
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}  -1.1992{col 37}{space 2} .1214277{col 48}{space 1}   -9.88{col 57}{space 3}0.000{col 65}{space 4}-1.437194{col 78}{space 3}-.9612064
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.158067{col 37}{space 2} .1204835{col 48}{space 1}   -9.61{col 57}{space 3}0.000{col 65}{space 4} -1.39421{col 78}{space 3}-.9219235
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.9405219{col 37}{space 2} .1258766{col 48}{space 1}   -7.47{col 57}{space 3}0.000{col 65}{space 4}-1.187235{col 78}{space 3}-.6938083
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-1.014981{col 37}{space 2} .1115601{col 48}{space 1}   -9.10{col 57}{space 3}0.000{col 65}{space 4}-1.233635{col 78}{space 3}-.7963272
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.9992414{col 37}{space 2} .1111618{col 48}{space 1}   -8.99{col 57}{space 3}0.000{col 65}{space 4}-1.217115{col 78}{space 3}-.7813683
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 14}turnoverb {c |}{col 25}{res}{space 2} -3.40671{col 37}{space 2} 2.504138{col 48}{space 1}   -1.36{col 57}{space 3}0.174{col 65}{space 4}-8.314731{col 78}{space 3} 1.501311
{txt}{space 1}itsepabove15_los15pctb {c |}{col 25}{res}{space 2}-.5388623{col 37}{space 2} 3.283947{col 48}{space 1}   -0.16{col 57}{space 3}0.870{col 65}{space 4}-6.975279{col 78}{space 3} 5.897555
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8867234{col 37}{space 2} .3860794{col 48}{space 1}    2.30{col 57}{space 3}0.022{col 65}{space 4} .1300217{col 78}{space 3} 1.643425
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .4185224{col 37}{space 2} .3124037{col 48}{space 1}    1.34{col 57}{space 3}0.180{col 65}{space 4}-.1937777{col 78}{space 3} 1.030822
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0073271{col 37}{space 2} .0615239{col 48}{space 1}    0.12{col 57}{space 3}0.905{col 65}{space 4}-.1132576{col 78}{space 3} .1279118
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0046607{col 37}{space 2} .0025937{col 48}{space 1}   -1.80{col 57}{space 3}0.072{col 65}{space 4}-.0097444{col 78}{space 3} .0004229
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-.6705202{col 37}{space 2} 1.438405{col 48}{space 1}   -0.47{col 57}{space 3}0.641{col 65}{space 4}-3.489743{col 78}{space 3} 2.148702
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-.9918512{col 37}{space 2} .3448887{col 48}{space 1}   -2.88{col 57}{space 3}0.004{col 65}{space 4}-1.667821{col 78}{space 3}-.3158819
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0294466{col 37}{space 2} .0611303{col 48}{space 1}   -0.48{col 57}{space 3}0.630{col 65}{space 4}-.1492598{col 78}{space 3} .0903666
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2}-.0191789{col 37}{space 2} .3828309{col 48}{space 1}   -0.05{col 57}{space 3}0.960{col 65}{space 4}-.7695137{col 78}{space 3} .7311559
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2}  .407628{col 37}{space 2} .5306603{col 48}{space 1}    0.77{col 57}{space 3}0.442{col 65}{space 4}-.6324472{col 78}{space 3} 1.447703
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .9234205{col 37}{space 2} .3391924{col 48}{space 1}    2.72{col 57}{space 3}0.006{col 65}{space 4} .2586157{col 78}{space 3} 1.588225
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est1{txt} stored)

{com}. 
. * Aggregate Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(turnover) 
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 4}turnover {c |}{col 14}{res}{space 2} .5781242{col 26}{space 2}  .341838{col 37}{space 1}    1.69{col 46}{space 3}0.091{col 54}{space 4}-.0918659{col 67}{space 3} 1.248114
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Aggregate Turnover Rate
. margins, at(turnover=($turnover_p10 $turnover_p90)) pwcompare(effects)
{res}
{txt}{col 1}Pairwise comparisons of predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.01}}
{lalign 7:2._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.09}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}   Contrast{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} .0464303{col 26}{space 2} .0275257{col 37}{space 1}    1.69{col 46}{space 3}0.092{col 54}{space 4}-.0075191{col 67}{space 3} .1003797
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3A] Unconditional Aggregate Turnover Effects on Project Delays (H1)
. margins, at(turnover=(0(0.01)0.15)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:0}}
{lalign 8:2._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.01}}
{lalign 8:3._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.02}}
{lalign 8:4._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.03}}
{lalign 8:5._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.04}}
{lalign 8:6._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.05}}
{lalign 8:7._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.06}}
{lalign 8:8._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.07}}
{lalign 8:9._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.08}}
{lalign 8:10._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.09}}
{lalign 8:11._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.1}}
{lalign 8:12._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.11}}
{lalign 8:13._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.12}}
{lalign 8:14._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.13}}
{lalign 8:15._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.14}}
{lalign 8:16._at: }{space 0}{lalign 8:turnover} = {res:{ralign 3:.15}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2694153{col 26}{space 2} .0191881{col 37}{space 1}   14.04{col 46}{space 3}0.000{col 54}{space 4} .2318072{col 67}{space 3} .3070233
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .2749651{col 26}{space 2} .0173361{col 37}{space 1}   15.86{col 46}{space 3}0.000{col 54}{space 4}  .240987{col 67}{space 3} .3089432
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .2805737{col 26}{space 2} .0158625{col 37}{space 1}   17.69{col 46}{space 3}0.000{col 54}{space 4} .2494838{col 67}{space 3} .3116635
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2862401{col 26}{space 2}  .014926{col 37}{space 1}   19.18{col 46}{space 3}0.000{col 54}{space 4} .2569858{col 67}{space 3} .3154945
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2919635{col 26}{space 2} .0146763{col 37}{space 1}   19.89{col 46}{space 3}0.000{col 54}{space 4} .2631985{col 67}{space 3} .3207284
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .2977426{col 26}{space 2} .0151914{col 37}{space 1}   19.60{col 46}{space 3}0.000{col 54}{space 4}  .267968{col 67}{space 3} .3275173
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .3035765{col 26}{space 2} .0164394{col 37}{space 1}   18.47{col 46}{space 3}0.000{col 54}{space 4}  .271356{col 67}{space 3} .3357971
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .3094641{col 26}{space 2} .0183056{col 37}{space 1}   16.91{col 46}{space 3}0.000{col 54}{space 4} .2735859{col 67}{space 3} .3453423
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .3154041{col 26}{space 2}  .020653{col 37}{space 1}   15.27{col 46}{space 3}0.000{col 54}{space 4}  .274925{col 67}{space 3} .3558832
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .3213954{col 26}{space 2} .0233625{col 37}{space 1}   13.76{col 46}{space 3}0.000{col 54}{space 4} .2756057{col 67}{space 3} .3671851
{txt}{space 9}11  {c |}{col 14}{res}{space 2} .3274367{col 26}{space 2} .0263445{col 37}{space 1}   12.43{col 46}{space 3}0.000{col 54}{space 4} .2758024{col 67}{space 3} .3790709
{txt}{space 9}12  {c |}{col 14}{res}{space 2} .3335267{col 26}{space 2}  .029535{col 37}{space 1}   11.29{col 46}{space 3}0.000{col 54}{space 4}  .275639{col 67}{space 3} .3914143
{txt}{space 9}13  {c |}{col 14}{res}{space 2}  .339664{col 26}{space 2} .0328892{col 37}{space 1}   10.33{col 46}{space 3}0.000{col 54}{space 4} .2752023{col 67}{space 3} .4041258
{txt}{space 9}14  {c |}{col 14}{res}{space 2} .3458474{col 26}{space 2} .0363751{col 37}{space 1}    9.51{col 46}{space 3}0.000{col 54}{space 4} .2745535{col 67}{space 3} .4171414
{txt}{space 9}15  {c |}{col 14}{res}{space 2} .3520754{col 26}{space 2} .0399695{col 37}{space 1}    8.81{col 46}{space 3}0.000{col 54}{space 4} .2737365{col 67}{space 3} .4304143
{txt}{space 9}16  {c |}{col 14}{res}{space 2} .3583465{col 26}{space 2}  .043655{col 37}{space 1}    8.21{col 46}{space 3}0.000{col 54}{space 4} .2727843{col 67}{space 3} .4439088
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30))  recastci(rarea) ///
> xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ///
> legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Aggregate Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3A. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Aggregate Turnover Model [MODEL 1]{c )-}", size(12pt) margin(medium)) ysc(r(0.1 0.5)) ///
> addplot(histogram turnover if turnover<=0.15, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:turnover}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3A.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3A.04-29-2025.gph} saved

{com}. *
. *
. *generate average cost per project variable for substantive interpretation
. gen cost_perproj = projects_size/projects_no
{txt}
{com}. sum cost_perproj if e(sample)

{txt}    Variable {c |}        Obs        Mean    Std. dev.       Min        Max
{hline 13}{c +}{hline 57}
cost_perproj {c |}{res}        540    7.194648    56.62391     .00025   974.9824
{txt}
{com}. * mean: 7.194648
. di 7.20*5.34
{res}38.448
{txt}
{com}. 
. *******************************************************************************************************************************************************************************************************************************
. 
. * [M2, FIGURE 4A] H2: Turnover X Service Length
. eststo: xtgee delay_pct c.turnover##c.itsepabove15_los15pct i.cio5 activityno_permil_pct cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnoverb itsepabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   dmeratio_totalspending dmeratio_totalspendingb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.
note: {bf:dmeratio_totalspending} omitted because of collinearity.
note: {bf:dmeratio_totalspendingb} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.64098873
{txt}Iteration 2:  Tolerance = {res}.24609791
{txt}Iteration 3:  Tolerance = {res}.26869691
{txt}Iteration 4:  Tolerance = {res}.19799009
{txt}Iteration 5:  Tolerance = {res}.13400438
{txt}Iteration 6:  Tolerance = {res}.082853
{txt}Iteration 7:  Tolerance = {res}.04796745
{txt}Iteration 8:  Tolerance = {res}.02656913
{txt}Iteration 9:  Tolerance = {res}.01431093
{txt}Iteration 10: Tolerance = {res}.00757593
{txt}Iteration 11: Tolerance = {res}.00396704
{txt}Iteration 12: Tolerance = {res}.0020625
{txt}Iteration 13: Tolerance = {res}.00106704
{txt}Iteration 14: Tolerance = {res}.00055005
{txt}Iteration 15: Tolerance = {res}.00028276
{txt}Iteration 16: Tolerance = {res}.00014504
{txt}Iteration 17: Tolerance = {res}.00007426
{txt}Iteration 18: Tolerance = {res}.00003796
{txt}Iteration 19: Tolerance = {res}.00001938
{txt}Iteration 20: Tolerance = {res}9.883e-06
{txt}Iteration 21: Tolerance = {res}5.035e-06
{txt}Iteration 22: Tolerance = {res}2.563e-06
{txt}Iteration 23: Tolerance = {res}1.304e-06
{txt}Iteration 24: Tolerance = {res}6.626e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:355.63}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 100:(Std. err. adjusted for clustering on {res:a_id})}
{hline 35}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 36}{c |}{col 48}  Semirobust
{col 1}                         delay_pct{col 36}{c |} Coefficient{col 48}  std. err.{col 60}      z{col 68}   P>|z|{col 76}     [95% con{col 89}f. interval]
{hline 35}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}turnover {c |}{col 36}{res}{space 2} 1.943289{col 48}{space 2} 1.334455{col 59}{space 1}    1.46{col 68}{space 3}0.145{col 76}{space 4}-.6721938{col 89}{space 3} 4.558772
{txt}{space 13}itsepabove15_los15pct {c |}{col 36}{res}{space 2}-.3915125{col 48}{space 2}  .830987{col 59}{space 1}   -0.47{col 68}{space 3}0.638{col 76}{space 4}-2.020217{col 89}{space 3} 1.237192
{txt}{space 34} {c |}
c.turnover#c.itsepabove15_los15pct {c |}{col 36}{res}{space 2}-.2120007{col 48}{space 2} 5.532002{col 59}{space 1}   -0.04{col 68}{space 3}0.969{col 76}{space 4}-11.05452{col 89}{space 3} 10.63052
{txt}{space 34} {c |}
{space 28}1.cio5 {c |}{col 36}{res}{space 2}-.0461953{col 48}{space 2} .0955213{col 59}{space 1}   -0.48{col 68}{space 3}0.629{col 76}{space 4}-.2334137{col 89}{space 3} .1410231
{txt}{space 13}activityno_permil_pct {c |}{col 36}{res}{space 2} .2312146{col 48}{space 2} .2274963{col 59}{space 1}    1.02{col 68}{space 3}0.309{col 76}{space 4}-.2146699{col 89}{space 3}  .677099
{txt}{space 24}cio_tenure {c |}{col 36}{res}{space 2}-.0116607{col 48}{space 2} .0270131{col 59}{space 1}   -0.43{col 68}{space 3}0.666{col 76}{space 4}-.0646054{col 89}{space 3} .0412841
{txt}{space 27}lnitemp {c |}{col 36}{res}{space 2}-.3918195{col 48}{space 2} .3132358{col 59}{space 1}   -1.25{col 68}{space 3}0.211{col 76}{space 4} -1.00575{col 89}{space 3} .2221114
{txt}{space 24}lnprojsize {c |}{col 36}{res}{space 2} .0361272{col 48}{space 2} .0299684{col 59}{space 1}    1.21{col 68}{space 3}0.228{col 76}{space 4}-.0226099{col 89}{space 3} .0948642
{txt}{space 23}projects_no {c |}{col 36}{res}{space 2} .0013155{col 48}{space 2} .0016668{col 59}{space 1}    0.79{col 68}{space 3}0.430{col 76}{space 4}-.0019515{col 89}{space 3} .0045824
{txt}{space 27}accrate {c |}{col 36}{res}{space 2} .0593889{col 48}{space 2} .5480048{col 59}{space 1}    0.11{col 68}{space 3}0.914{col 76}{space 4}-1.014681{col 89}{space 3} 1.133459
{txt}{space 19}newprojects_pct {c |}{col 36}{res}{space 2}-.8312811{col 48}{space 2} .1722908{col 59}{space 1}   -4.82{col 68}{space 3}0.000{col 76}{space 4}-1.168965{col 89}{space 3}-.4935973
{txt}{space 32}y1 {c |}{col 36}{res}{space 2}-1.252211{col 48}{space 2}  .101042{col 59}{space 1}  -12.39{col 68}{space 3}0.000{col 76}{space 4} -1.45025{col 89}{space 3}-1.054173
{txt}{space 32}y2 {c |}{col 36}{res}{space 2}-1.199306{col 48}{space 2} .1213729{col 59}{space 1}   -9.88{col 68}{space 3}0.000{col 76}{space 4}-1.437192{col 89}{space 3}-.9614194
{txt}{space 32}y3 {c |}{col 36}{res}{space 2}-1.157969{col 48}{space 2} .1203975{col 59}{space 1}   -9.62{col 68}{space 3}0.000{col 76}{space 4}-1.393944{col 89}{space 3}-.9219945
{txt}{space 32}y4 {c |}{col 36}{res}{space 2}-.9405418{col 48}{space 2} .1255046{col 59}{space 1}   -7.49{col 68}{space 3}0.000{col 76}{space 4}-1.186526{col 89}{space 3}-.6945574
{txt}{space 32}y5 {c |}{col 36}{res}{space 2} -1.01489{col 48}{space 2} .1116378{col 59}{space 1}   -9.09{col 68}{space 3}0.000{col 76}{space 4}-1.233696{col 89}{space 3}-.7960839
{txt}{space 32}y6 {c |}{col 36}{res}{space 2}-.9992695{col 48}{space 2} .1115593{col 59}{space 1}   -8.96{col 68}{space 3}0.000{col 76}{space 4}-1.217922{col 89}{space 3}-.7806172
{txt}{space 32}y7 {c |}{col 36}{res}{space 2}        0{col 48}{txt}  (omitted)
{space 25}turnoverb {c |}{col 36}{res}{space 2}-3.408889{col 48}{space 2} 2.502439{col 59}{space 1}   -1.36{col 68}{space 3}0.173{col 76}{space 4} -8.31358{col 89}{space 3} 1.495802
{txt}{space 12}itsepabove15_los15pctb {c |}{col 36}{res}{space 2}-.5384863{col 48}{space 2} 3.285336{col 59}{space 1}   -0.16{col 68}{space 3}0.870{col 76}{space 4}-6.977627{col 89}{space 3} 5.900654
{txt}{space 12}activityno_permil_pctb {c |}{col 36}{res}{space 2} .8875859{col 48}{space 2} .3939237{col 59}{space 1}    2.25{col 68}{space 3}0.024{col 76}{space 4} .1155095{col 89}{space 3} 1.659662
{txt}{space 26}lnitempb {c |}{col 36}{res}{space 2} .4199814{col 48}{space 2} .3173302{col 59}{space 1}    1.32{col 68}{space 3}0.186{col 76}{space 4}-.2019745{col 89}{space 3} 1.041937
{txt}{space 23}lnprojsizeb {c |}{col 36}{res}{space 2} .0075317{col 48}{space 2} .0634262{col 59}{space 1}    0.12{col 68}{space 3}0.905{col 76}{space 4}-.1167813{col 89}{space 3} .1318447
{txt}{space 22}projects_nob {c |}{col 36}{res}{space 2}-.0046584{col 48}{space 2} .0026002{col 59}{space 1}   -1.79{col 68}{space 3}0.073{col 76}{space 4}-.0097546{col 89}{space 3} .0004378
{txt}{space 26}accrateb {c |}{col 36}{res}{space 2}-.6762851{col 48}{space 2} 1.453181{col 59}{space 1}   -0.47{col 68}{space 3}0.642{col 76}{space 4}-3.524467{col 89}{space 3} 2.171897
{txt}{space 18}newprojects_pctb {c |}{col 36}{res}{space 2}-.9914616{col 48}{space 2} .3448194{col 59}{space 1}   -2.88{col 68}{space 3}0.004{col 76}{space 4}-1.667295{col 89}{space 3} -.315628
{txt}{space 23}cio_tenureb {c |}{col 36}{res}{space 2}-.0294031{col 48}{space 2} .0610399{col 59}{space 1}   -0.48{col 68}{space 3}0.630{col 76}{space 4} -.149039{col 89}{space 3} .0902329
{txt}{space 12}dmeratio_totalspending {c |}{col 36}{res}{space 2} -.018856{col 48}{space 2} .3821921{col 59}{space 1}   -0.05{col 68}{space 3}0.961{col 76}{space 4}-.7679387{col 89}{space 3} .7302267
{txt}{space 11}dmeratio_totalspendingb {c |}{col 36}{res}{space 2}  .407442{col 48}{space 2} .5303492{col 59}{space 1}    0.77{col 68}{space 3}0.442{col 76}{space 4}-.6320232{col 89}{space 3} 1.446907
{txt}{space 12}dmeratio_totalspending {c |}{col 36}{res}{space 2}        0{col 48}{txt}  (omitted)
{space 11}dmeratio_totalspendingb {c |}{col 36}{res}{space 2}        0{col 48}{txt}  (omitted)
{space 29}_cons {c |}{col 36}{res}{space 2} .9223352{col 48}{space 2} .3433599{col 59}{space 1}    2.69{col 68}{space 3}0.007{col 76}{space 4} .2493622{col 89}{space 3} 1.595308
{txt}{hline 35}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est2{txt} stored)

{com}. 
. * Aggregate Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins, dydx(turnover) at(itsepabove15_los15pct=($los_p10 $los_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 3:.02}}
{lalign 7:2._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 3:.14}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2}-.0216567{col 26}{space 2}  .191496{col 37}{space 1}   -0.11{col 46}{space 3}0.910{col 54}{space 4} -.396982{col 67}{space 3} .3536686
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 4A] Conditional Turnover Effects by Departures' Service Length (H2)
. margins, dydx(turnover) at(itsepabove15_los15pct=(0(0.025)0.2))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:0}}
{lalign 7:2._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.025}}
{lalign 7:3._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.05}}
{lalign 7:4._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.075}}
{lalign 7:5._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.1}}
{lalign 7:6._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.125}}
{lalign 7:7._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.15}}
{lalign 7:8._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.175}}
{lalign 7:9._at: }{space 0}{lalign 16:itsepabove15_l~t} = {res:{ralign 4:.2}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .5968685{col 26}{space 2} .4132227{col 37}{space 1}    1.44{col 46}{space 3}0.149{col 54}{space 4}-.2130332{col 67}{space 3}  1.40677
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .5923965{col 26}{space 2} .3904142{col 37}{space 1}    1.52{col 46}{space 3}0.129{col 54}{space 4}-.1728014{col 67}{space 3} 1.357594
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .5879087{col 26}{space 2} .3708652{col 37}{space 1}    1.59{col 46}{space 3}0.113{col 54}{space 4}-.1389737{col 67}{space 3} 1.314791
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .5834059{col 26}{space 2} .3549712{col 37}{space 1}    1.64{col 46}{space 3}0.100{col 54}{space 4}-.1123248{col 67}{space 3} 1.279137
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .5788892{col 26}{space 2} .3430918{col 37}{space 1}    1.69{col 46}{space 3}0.092{col 54}{space 4}-.0935584{col 67}{space 3} 1.251337
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .5743594{col 26}{space 2} .3355011{col 37}{space 1}    1.71{col 46}{space 3}0.087{col 54}{space 4}-.0832108{col 67}{space 3} 1.231929
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .5698172{col 26}{space 2} .3323386{col 37}{space 1}    1.71{col 46}{space 3}0.086{col 54}{space 4}-.0815545{col 67}{space 3} 1.221189
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .5652637{col 26}{space 2} .3335763{col 37}{space 1}    1.69{col 46}{space 3}0.090{col 54}{space 4}-.0885337{col 67}{space 3} 1.219061
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .5606997{col 26}{space 2} .3390144{col 37}{space 1}    1.65{col 46}{space 3}0.098{col 54}{space 4}-.1037562{col 67}{space 3} 1.225156
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> xlabel(0 "0" 0.1 "10" 0.2 "20" ,format(%9.0f) ) ylabel(,format(%9.1f) angle(0)) ///
> legend(off) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departing Employees' Service Length (>15 years, %)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 4A. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Aggregate Turnover Model [MODEL 2]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram itsepabove15_los15pct if itsepabove15_los15pct<=0.2 & turnover>0, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.1 "10" 0.2 "20" ,format(%9.0f)) ylabel(,format(%9.1f) angle(0) ) ylabel(0 "0" 5 "5" 10 "10" 15 "15" 20 "20",format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:itsepabove15_los15pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4A.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4A.04-29-2025.gph} saved

{com}. 
. *******************************************************************************************************************************************************************************************************************************
. 
. * [M3, FIGURE 5A] H3: Turnover X Agency Centralization
. eststo: xtgee delay_pct c.turnover##i.cio5 c.itsepabove15_los15pct  activityno_permil_pct cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnoverb itsepabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}1.7159053
{txt}Iteration 2:  Tolerance = {res}.0620385
{txt}Iteration 3:  Tolerance = {res}.02287123
{txt}Iteration 4:  Tolerance = {res}.00530867
{txt}Iteration 5:  Tolerance = {res}.00222223
{txt}Iteration 6:  Tolerance = {res}.00094409
{txt}Iteration 7:  Tolerance = {res}.00046127
{txt}Iteration 8:  Tolerance = {res}.00023162
{txt}Iteration 9:  Tolerance = {res}.00011099
{txt}Iteration 10: Tolerance = {res}.00005134
{txt}Iteration 11: Tolerance = {res}.00002307
{txt}Iteration 12: Tolerance = {res}.0000101
{txt}Iteration 13: Tolerance = {res}4.315e-06
{txt}Iteration 14: Tolerance = {res}1.808e-06
{txt}Iteration 15: Tolerance = {res}7.924e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:345.74}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 15}turnover {c |}{col 25}{res}{space 2} 2.441501{col 37}{space 2} 1.184479{col 48}{space 1}    2.06{col 57}{space 3}0.039{col 65}{space 4} .1199653{col 78}{space 3} 4.763036
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2} .0836051{col 37}{space 2} .1261534{col 48}{space 1}    0.66{col 57}{space 3}0.508{col 65}{space 4}-.1636511{col 78}{space 3} .3308612
{txt}{space 23} {c |}
{space 8}cio5#c.turnover {c |}
{space 21}1  {c |}{col 25}{res}{space 2}-2.927227{col 37}{space 2}  1.38038{col 48}{space 1}   -2.12{col 57}{space 3}0.034{col 65}{space 4}-5.632722{col 78}{space 3}-.2217321
{txt}{space 23} {c |}
{space 2}itsepabove15_los15pct {c |}{col 25}{res}{space 2}-.4096446{col 37}{space 2} .5745045{col 48}{space 1}   -0.71{col 57}{space 3}0.476{col 65}{space 4}-1.535653{col 78}{space 3} .7163635
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2}   .24412{col 37}{space 2} .2198133{col 48}{space 1}    1.11{col 57}{space 3}0.267{col 65}{space 4}-.1867062{col 78}{space 3} .6749462
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0143498{col 37}{space 2} .0265332{col 48}{space 1}   -0.54{col 57}{space 3}0.589{col 65}{space 4}-.0663539{col 78}{space 3} .0376544
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.3597761{col 37}{space 2} .3070592{col 48}{space 1}   -1.17{col 57}{space 3}0.241{col 65}{space 4} -.961601{col 78}{space 3} .2420488
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0412083{col 37}{space 2} .0283722{col 48}{space 1}    1.45{col 57}{space 3}0.146{col 65}{space 4}-.0144002{col 78}{space 3} .0968169
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0011599{col 37}{space 2} .0016699{col 48}{space 1}    0.69{col 57}{space 3}0.487{col 65}{space 4} -.002113{col 78}{space 3} .0044328
{txt}{space 16}accrate {c |}{col 25}{res}{space 2}-.0778089{col 37}{space 2} .5521717{col 48}{space 1}   -0.14{col 57}{space 3}0.888{col 65}{space 4}-1.160046{col 78}{space 3} 1.004428
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8272568{col 37}{space 2} .1697059{col 48}{space 1}   -4.87{col 57}{space 3}0.000{col 65}{space 4}-1.159874{col 78}{space 3}-.4946394
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.256248{col 37}{space 2} .1012043{col 48}{space 1}  -12.41{col 57}{space 3}0.000{col 65}{space 4}-1.454605{col 78}{space 3}-1.057891
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.195197{col 37}{space 2}  .121784{col 48}{space 1}   -9.81{col 57}{space 3}0.000{col 65}{space 4}-1.433889{col 78}{space 3} -.956505
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.154663{col 37}{space 2} .1196778{col 48}{space 1}   -9.65{col 57}{space 3}0.000{col 65}{space 4}-1.389227{col 78}{space 3}-.9200983
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.9610799{col 37}{space 2} .1270993{col 48}{space 1}   -7.56{col 57}{space 3}0.000{col 65}{space 4} -1.21019{col 78}{space 3}-.7119698
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-1.006997{col 37}{space 2} .1112196{col 48}{space 1}   -9.05{col 57}{space 3}0.000{col 65}{space 4}-1.224984{col 78}{space 3}-.7890109
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.9905568{col 37}{space 2} .1094901{col 48}{space 1}   -9.05{col 57}{space 3}0.000{col 65}{space 4}-1.205154{col 78}{space 3}-.7759601
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 14}turnoverb {c |}{col 25}{res}{space 2}-3.503196{col 37}{space 2} 2.542083{col 48}{space 1}   -1.38{col 57}{space 3}0.168{col 65}{space 4}-8.485587{col 78}{space 3} 1.479195
{txt}{space 1}itsepabove15_los15pctb {c |}{col 25}{res}{space 2}-.4742004{col 37}{space 2} 3.320472{col 48}{space 1}   -0.14{col 57}{space 3}0.886{col 65}{space 4}-6.982207{col 78}{space 3} 6.033806
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8843127{col 37}{space 2}  .383203{col 48}{space 1}    2.31{col 57}{space 3}0.021{col 65}{space 4} .1332485{col 78}{space 3} 1.635377
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .3878949{col 37}{space 2} .3117924{col 48}{space 1}    1.24{col 57}{space 3}0.213{col 65}{space 4}-.2232069{col 78}{space 3} .9989967
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0047098{col 37}{space 2} .0609567{col 48}{space 1}    0.08{col 57}{space 3}0.938{col 65}{space 4}-.1147631{col 78}{space 3} .1241826
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0047844{col 37}{space 2} .0025741{col 48}{space 1}   -1.86{col 57}{space 3}0.063{col 65}{space 4}-.0098296{col 78}{space 3} .0002608
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2} -.558614{col 37}{space 2} 1.461852{col 48}{space 1}   -0.38{col 57}{space 3}0.702{col 65}{space 4}-3.423792{col 78}{space 3} 2.306564
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-.9669162{col 37}{space 2}  .347247{col 48}{space 1}   -2.78{col 57}{space 3}0.005{col 65}{space 4}-1.647508{col 78}{space 3}-.2863246
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0231911{col 37}{space 2} .0621011{col 48}{space 1}   -0.37{col 57}{space 3}0.709{col 65}{space 4} -.144907{col 78}{space 3} .0985248
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2}-.0323878{col 37}{space 2} .3785425{col 48}{space 1}   -0.09{col 57}{space 3}0.932{col 65}{space 4}-.7743174{col 78}{space 3} .7095418
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .4485454{col 37}{space 2}  .526554{col 48}{space 1}    0.85{col 57}{space 3}0.394{col 65}{space 4}-.5834814{col 78}{space 3} 1.480572
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .8680855{col 37}{space 2} .3502243{col 48}{space 1}    2.48{col 57}{space 3}0.013{col 65}{space 4} .1816584{col 78}{space 3} 1.554513
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est3{txt} stored)

{com}. 
. * Aggregate Turnover Rate: Marginal Effect Differentials Based on Agency Centralization
. margins , dydx(turnover) over(cio5) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 8}cio5 {c |}
{space 5}1 vs 0  {c |}{col 14}{res}{space 2}-.8836369{col 26}{space 2} .4108515{col 37}{space 1}   -2.15{col 46}{space 3}0.031{col 54}{space 4}-1.688891{col 67}{space 3}-.0783827
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 5A] Conditional Turnover Effects by Agency Centralization (H3)
. margins , dydx(turnover) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 8}cio5 {c |}
{space 10}0  {c |}{col 14}{res}{space 2} .7391057{col 26}{space 2}  .360862{col 37}{space 1}    2.05{col 46}{space 3}0.041{col 54}{space 4} .0318291{col 67}{space 3} 1.446382
{txt}{space 10}1  {c |}{col 14}{res}{space 2}-.1445312{col 26}{space 2} .4341375{col 37}{space 1}   -0.33{col 46}{space 3}0.739{col 54}{space 4} -.995425{col 67}{space 3} .7063626
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(black)) ciopt(color(black%60))  yline(0, lcolor(red) lpattern(shortdash)) ///
> xlabel (-1 " " 0 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel(,format(%9.1f) angle(0))   ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5A. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Aggregate Turnover Model [MODEL 3]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5A.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5A.04-29-2025.gph} saved

{com}. 
. ******************************************************************************************************************************************************************************************************************************
. 
. * [M4, FIGURE 6A] H4: Turnover X Task Standardization
. eststo: xtgee delay_pct c.turnover##c.activityno_permil_pct i.cio5 c.itsepabove15_los15pct   cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnoverb itsepabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}1.5375442
{txt}Iteration 2:  Tolerance = {res}.07316186
{txt}Iteration 3:  Tolerance = {res}.02934857
{txt}Iteration 4:  Tolerance = {res}.00567348
{txt}Iteration 5:  Tolerance = {res}.00239801
{txt}Iteration 6:  Tolerance = {res}.00115004
{txt}Iteration 7:  Tolerance = {res}.00055393
{txt}Iteration 8:  Tolerance = {res}.00028248
{txt}Iteration 9:  Tolerance = {res}.00013905
{txt}Iteration 10: Tolerance = {res}.00006695
{txt}Iteration 11: Tolerance = {res}.00003161
{txt}Iteration 12: Tolerance = {res}.0000147
{txt}Iteration 13: Tolerance = {res}6.741e-06
{txt}Iteration 14: Tolerance = {res}3.054e-06
{txt}Iteration 15: Tolerance = {res}1.368e-06
{txt}Iteration 16: Tolerance = {res}6.067e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:376.62}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 100:(Std. err. adjusted for clustering on {res:a_id})}
{hline 35}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 36}{c |}{col 48}  Semirobust
{col 1}                         delay_pct{col 36}{c |} Coefficient{col 48}  std. err.{col 60}      z{col 68}   P>|z|{col 76}     [95% con{col 89}f. interval]
{hline 35}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}turnover {c |}{col 36}{res}{space 2} 2.564942{col 48}{space 2} 1.514777{col 59}{space 1}    1.69{col 68}{space 3}0.090{col 76}{space 4} -.403966{col 89}{space 3}  5.53385
{txt}{space 13}activityno_permil_pct {c |}{col 36}{res}{space 2} .3135873{col 48}{space 2} .2738797{col 59}{space 1}    1.14{col 68}{space 3}0.252{col 76}{space 4}-.2232069{col 89}{space 3} .8503816
{txt}{space 34} {c |}
c.turnover#c.activityno_permil_pct {c |}{col 36}{res}{space 2}-1.748868{col 48}{space 2} 2.372587{col 59}{space 1}   -0.74{col 68}{space 3}0.461{col 76}{space 4}-6.399053{col 89}{space 3} 2.901318
{txt}{space 34} {c |}
{space 28}1.cio5 {c |}{col 36}{res}{space 2}-.0401273{col 48}{space 2} .0947293{col 59}{space 1}   -0.42{col 68}{space 3}0.672{col 76}{space 4}-.2257933{col 89}{space 3} .1455387
{txt}{space 13}itsepabove15_los15pct {c |}{col 36}{res}{space 2}-.4441194{col 48}{space 2} .5594393{col 59}{space 1}   -0.79{col 68}{space 3}0.427{col 76}{space 4}  -1.5406{col 89}{space 3} .6523614
{txt}{space 24}cio_tenure {c |}{col 36}{res}{space 2}-.0100711{col 48}{space 2} .0266881{col 59}{space 1}   -0.38{col 68}{space 3}0.706{col 76}{space 4}-.0623788{col 89}{space 3} .0422365
{txt}{space 27}lnitemp {c |}{col 36}{res}{space 2} -.389962{col 48}{space 2} .3067606{col 59}{space 1}   -1.27{col 68}{space 3}0.204{col 76}{space 4}-.9912017{col 89}{space 3} .2112777
{txt}{space 24}lnprojsize {c |}{col 36}{res}{space 2} .0363987{col 48}{space 2} .0294362{col 59}{space 1}    1.24{col 68}{space 3}0.216{col 76}{space 4}-.0212952{col 89}{space 3} .0940926
{txt}{space 23}projects_no {c |}{col 36}{res}{space 2}  .001423{col 48}{space 2} .0016678{col 59}{space 1}    0.85{col 68}{space 3}0.394{col 76}{space 4}-.0018459{col 89}{space 3} .0046919
{txt}{space 27}accrate {c |}{col 36}{res}{space 2} .0305251{col 48}{space 2} .5375438{col 59}{space 1}    0.06{col 68}{space 3}0.955{col 76}{space 4}-1.023041{col 89}{space 3} 1.084092
{txt}{space 19}newprojects_pct {c |}{col 36}{res}{space 2}-.8357648{col 48}{space 2} .1694516{col 59}{space 1}   -4.93{col 68}{space 3}0.000{col 76}{space 4}-1.167884{col 89}{space 3}-.5036458
{txt}{space 32}y1 {c |}{col 36}{res}{space 2} -1.25871{col 48}{space 2} .0991648{col 59}{space 1}  -12.69{col 68}{space 3}0.000{col 76}{space 4}-1.453069{col 89}{space 3} -1.06435
{txt}{space 32}y2 {c |}{col 36}{res}{space 2}-1.205121{col 48}{space 2} .1204831{col 59}{space 1}  -10.00{col 68}{space 3}0.000{col 76}{space 4}-1.441264{col 89}{space 3}-.9689789
{txt}{space 32}y3 {c |}{col 36}{res}{space 2}-1.165682{col 48}{space 2} .1194806{col 59}{space 1}   -9.76{col 68}{space 3}0.000{col 76}{space 4} -1.39986{col 89}{space 3}-.9315048
{txt}{space 32}y4 {c |}{col 36}{res}{space 2}-.9465984{col 48}{space 2} .1248702{col 59}{space 1}   -7.58{col 68}{space 3}0.000{col 76}{space 4}-1.191339{col 89}{space 3}-.7018574
{txt}{space 32}y5 {c |}{col 36}{res}{space 2}-1.022135{col 48}{space 2} .1106856{col 59}{space 1}   -9.23{col 68}{space 3}0.000{col 76}{space 4}-1.239074{col 89}{space 3}-.8051949
{txt}{space 32}y6 {c |}{col 36}{res}{space 2}-1.001387{col 48}{space 2} .1110627{col 59}{space 1}   -9.02{col 68}{space 3}0.000{col 76}{space 4}-1.219066{col 89}{space 3}-.7837079
{txt}{space 32}y7 {c |}{col 36}{res}{space 2}        0{col 48}{txt}  (omitted)
{space 25}turnoverb {c |}{col 36}{res}{space 2}-3.509427{col 48}{space 2} 2.537888{col 59}{space 1}   -1.38{col 68}{space 3}0.167{col 76}{space 4}-8.483596{col 89}{space 3} 1.464743
{txt}{space 12}itsepabove15_los15pctb {c |}{col 36}{res}{space 2}-.5029342{col 48}{space 2} 3.263885{col 59}{space 1}   -0.15{col 68}{space 3}0.878{col 76}{space 4}-6.900031{col 89}{space 3} 5.894163
{txt}{space 12}activityno_permil_pctb {c |}{col 36}{res}{space 2} .8857375{col 48}{space 2} .3852062{col 59}{space 1}    2.30{col 68}{space 3}0.021{col 76}{space 4} .1307471{col 89}{space 3} 1.640728
{txt}{space 26}lnitempb {c |}{col 36}{res}{space 2} .4192483{col 48}{space 2} .3120403{col 59}{space 1}    1.34{col 68}{space 3}0.179{col 76}{space 4}-.1923394{col 89}{space 3} 1.030836
{txt}{space 23}lnprojsizeb {c |}{col 36}{res}{space 2} .0065455{col 48}{space 2} .0616593{col 59}{space 1}    0.11{col 68}{space 3}0.915{col 76}{space 4}-.1143045{col 89}{space 3} .1273955
{txt}{space 22}projects_nob {c |}{col 36}{res}{space 2}-.0047909{col 48}{space 2}   .00258{col 59}{space 1}   -1.86{col 68}{space 3}0.063{col 76}{space 4}-.0098477{col 89}{space 3} .0002659
{txt}{space 26}accrateb {c |}{col 36}{res}{space 2}-.6720565{col 48}{space 2} 1.422823{col 59}{space 1}   -0.47{col 68}{space 3}0.637{col 76}{space 4}-3.460739{col 89}{space 3} 2.116626
{txt}{space 18}newprojects_pctb {c |}{col 36}{res}{space 2}-.9807763{col 48}{space 2} .3449643{col 59}{space 1}   -2.84{col 68}{space 3}0.004{col 76}{space 4}-1.656894{col 89}{space 3}-.3046586
{txt}{space 23}cio_tenureb {c |}{col 36}{res}{space 2}-.0292958{col 48}{space 2} .0618284{col 59}{space 1}   -0.47{col 68}{space 3}0.636{col 76}{space 4}-.1504772{col 89}{space 3} .0918857
{txt}{space 12}dmeratio_totalspending {c |}{col 36}{res}{space 2} .0067332{col 48}{space 2}  .382773{col 59}{space 1}    0.02{col 68}{space 3}0.986{col 76}{space 4}-.7434881{col 89}{space 3} .7569544
{txt}{space 11}dmeratio_totalspendingb {c |}{col 36}{res}{space 2} .3962258{col 48}{space 2} .5342375{col 59}{space 1}    0.74{col 68}{space 3}0.458{col 76}{space 4}-.6508604{col 89}{space 3} 1.443312
{txt}{space 29}_cons {c |}{col 36}{res}{space 2} .8917259{col 48}{space 2} .3467655{col 59}{space 1}    2.57{col 68}{space 3}0.010{col 76}{space 4}  .212078{col 89}{space 3} 1.571374
{txt}{hline 35}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est4{txt} stored)

{com}. 
. * Aggregate Turnover Rate: Marginal Effect Differentials Based on Task Standardization
. margins , dydx(turnover) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2}-.3290395{col 26}{space 2} .4979988{col 37}{space 1}   -0.66{col 46}{space 3}0.509{col 54}{space 4}-1.305099{col 67}{space 3} .6470202
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 6A] Conditional Turnover Effects by Task Standardization (H4)
. margins , dydx(turnover) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover     {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .7479578{col 26}{space 2} .4443157{col 37}{space 1}    1.68{col 46}{space 3}0.092{col 54}{space 4}-.1228849{col 67}{space 3}   1.6188
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .7060148{col 26}{space 2} .4054807{col 37}{space 1}    1.74{col 46}{space 3}0.082{col 54}{space 4}-.0887128{col 67}{space 3} 1.500742
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .6625698{col 26}{space 2} .3725397{col 37}{space 1}    1.78{col 46}{space 3}0.075{col 54}{space 4}-.0675947{col 67}{space 3} 1.392734
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .6176537{col 26}{space 2} .3484225{col 37}{space 1}    1.77{col 46}{space 3}0.076{col 54}{space 4}-.0652418{col 67}{space 3} 1.300549
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .5713012{col 26}{space 2}  .336348{col 37}{space 1}    1.70{col 46}{space 3}0.089{col 54}{space 4}-.0879288{col 67}{space 3} 1.230531
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .5235509{col 26}{space 2} .3389082{col 37}{space 1}    1.54{col 46}{space 3}0.122{col 54}{space 4} -.140697{col 67}{space 3} 1.187799
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .4744451{col 26}{space 2}  .357001{col 37}{space 1}    1.33{col 46}{space 3}0.184{col 54}{space 4}-.2252641{col 67}{space 3} 1.174154
{txt}{space 10}8  {c |}{col 14}{res}{space 2}   .42403{col 26}{space 2} .3895424{col 37}{space 1}    1.09{col 46}{space 3}0.276{col 54}{space 4}-.3394592{col 67}{space 3} 1.187519
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .3723553{col 26}{space 2} .4342144{col 37}{space 1}    0.86{col 46}{space 3}0.391{col 54}{space 4}-.4786893{col 67}{space 3}   1.2234
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .3194747{col 26}{space 2} .4884653{col 37}{space 1}    0.65{col 46}{space 3}0.513{col 54}{space 4}-.6378997{col 67}{space 3} 1.276849
{txt}{space 9}11  {c |}{col 14}{res}{space 2}  .265445{col 26}{space 2}  .550091{col 37}{space 1}    0.48{col 46}{space 3}0.629{col 54}{space 4}-.8127135{col 67}{space 3} 1.343604
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", size(11pt) margin(t=1)) ///
> title("{c -(}bf: Figure 6A. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Aggregate Turnover Model [MODEL 4]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0))  ylabel( ,format(%9.1f) angle(0) axis(2)) ytitle(, axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6A.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6A.04-29-2025.gph} saved

{com}. 
. ******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. **# [MODEL 5 - 8] [FIGURES 3D-6D] MANAGERIAL TURNOVER MODELS
. 
. * [M5, FIGURE 3D] H1: Managerial Turnover Baseline Effects
.  
. eststo: xtgee delay_pct c.turnover_gs13  gs13above15_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnover_gs13b gs13above15_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.34366411
{txt}Iteration 2:  Tolerance = {res}.16531215
{txt}Iteration 3:  Tolerance = {res}.05735119
{txt}Iteration 4:  Tolerance = {res}.02024504
{txt}Iteration 5:  Tolerance = {res}.0072762
{txt}Iteration 6:  Tolerance = {res}.00344846
{txt}Iteration 7:  Tolerance = {res}.00157089
{txt}Iteration 8:  Tolerance = {res}.00070307
{txt}Iteration 9:  Tolerance = {res}.00031209
{txt}Iteration 10: Tolerance = {res}.00013799
{txt}Iteration 11: Tolerance = {res}.00006089
{txt}Iteration 12: Tolerance = {res}.00002684
{txt}Iteration 13: Tolerance = {res}.00001183
{txt}Iteration 14: Tolerance = {res}5.210e-06
{txt}Iteration 15: Tolerance = {res}2.295e-06
{txt}Iteration 16: Tolerance = {res}1.011e-06
{txt}Iteration 17: Tolerance = {res}4.452e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:326.62}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 10}turnover_gs13 {c |}{col 25}{res}{space 2} .3806663{col 37}{space 2} 1.959147{col 48}{space 1}    0.19{col 57}{space 3}0.846{col 65}{space 4}-3.459191{col 78}{space 3} 4.220523
{txt}{space 3}gs13above15_los15pct {c |}{col 25}{res}{space 2}-.2457694{col 37}{space 2} 1.116133{col 48}{space 1}   -0.22{col 57}{space 3}0.826{col 65}{space 4}-2.433349{col 78}{space 3}  1.94181
{txt}{space 19}cio5 {c |}{col 25}{res}{space 2}-.0913788{col 37}{space 2} .0825891{col 48}{space 1}   -1.11{col 57}{space 3}0.269{col 65}{space 4}-.2532504{col 78}{space 3} .0704929
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2806764{col 37}{space 2} .1996753{col 48}{space 1}    1.41{col 57}{space 3}0.160{col 65}{space 4}-.1106799{col 78}{space 3} .6720328
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0126564{col 37}{space 2} .0244251{col 48}{space 1}   -0.52{col 57}{space 3}0.604{col 65}{space 4}-.0605287{col 78}{space 3} .0352158
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .1136231{col 37}{space 2} .2374016{col 48}{space 1}    0.48{col 57}{space 3}0.632{col 65}{space 4}-.3516756{col 78}{space 3} .5789217
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0787349{col 37}{space 2} .0293556{col 48}{space 1}    2.68{col 57}{space 3}0.007{col 65}{space 4} .0211991{col 78}{space 3} .1362708
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0011004{col 37}{space 2} .0017463{col 48}{space 1}    0.63{col 57}{space 3}0.529{col 65}{space 4}-.0023223{col 78}{space 3} .0045231
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .3978969{col 37}{space 2} .5458318{col 48}{space 1}    0.73{col 57}{space 3}0.466{col 65}{space 4}-.6719138{col 78}{space 3} 1.467708
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8155448{col 37}{space 2} .1658885{col 48}{space 1}   -4.92{col 57}{space 3}0.000{col 65}{space 4} -1.14068{col 78}{space 3}-.4904094
{txt}{space 21}y1 {c |}{col 25}{res}{space 2} -1.19222{col 37}{space 2} .1043075{col 48}{space 1}  -11.43{col 57}{space 3}0.000{col 65}{space 4}-1.396659{col 78}{space 3}-.9877811
{txt}{space 21}y2 {c |}{col 25}{res}{space 2} -1.17825{col 37}{space 2} .1262516{col 48}{space 1}   -9.33{col 57}{space 3}0.000{col 65}{space 4}-1.425698{col 78}{space 3}-.9308014
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.120458{col 37}{space 2}  .118142{col 48}{space 1}   -9.48{col 57}{space 3}0.000{col 65}{space 4}-1.352012{col 78}{space 3}-.8889037
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.8687631{col 37}{space 2} .1199291{col 48}{space 1}   -7.24{col 57}{space 3}0.000{col 65}{space 4} -1.10382{col 78}{space 3}-.6337064
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-1.000708{col 37}{space 2} .1099032{col 48}{space 1}   -9.11{col 57}{space 3}0.000{col 65}{space 4}-1.216114{col 78}{space 3}-.7853013
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8996495{col 37}{space 2}  .103031{col 48}{space 1}   -8.73{col 57}{space 3}0.000{col 65}{space 4}-1.101587{col 78}{space 3}-.6977125
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 9}turnover_gs13b {c |}{col 25}{res}{space 2}-.0246828{col 37}{space 2} 3.248779{col 48}{space 1}   -0.01{col 57}{space 3}0.994{col 65}{space 4}-6.392173{col 78}{space 3} 6.342808
{txt}{space 2}gs13above15_los15pctb {c |}{col 25}{res}{space 2}-6.544529{col 37}{space 2} 2.282702{col 48}{space 1}   -2.87{col 57}{space 3}0.004{col 65}{space 4}-11.01854{col 78}{space 3}-2.070515
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9875427{col 37}{space 2} .4072543{col 48}{space 1}    2.42{col 57}{space 3}0.015{col 65}{space 4} .1893389{col 78}{space 3} 1.785747
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2}-.1053302{col 37}{space 2} .2427786{col 48}{space 1}   -0.43{col 57}{space 3}0.664{col 65}{space 4}-.5811674{col 78}{space 3} .3705071
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0008032{col 37}{space 2} .0640053{col 48}{space 1}    0.01{col 57}{space 3}0.990{col 65}{space 4}-.1246449{col 78}{space 3} .1262513
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0054453{col 37}{space 2} .0032912{col 48}{space 1}   -1.65{col 57}{space 3}0.098{col 65}{space 4}-.0118959{col 78}{space 3} .0010053
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}   -.9373{col 37}{space 2} 1.226615{col 48}{space 1}   -0.76{col 57}{space 3}0.445{col 65}{space 4}-3.341421{col 78}{space 3} 1.466821
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.009255{col 37}{space 2} .3391646{col 48}{space 1}   -2.98{col 57}{space 3}0.003{col 65}{space 4}-1.674005{col 78}{space 3}-.3445045
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0348168{col 37}{space 2}  .055484{col 48}{space 1}   -0.63{col 57}{space 3}0.530{col 65}{space 4}-.1435634{col 78}{space 3} .0739298
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .3266718{col 37}{space 2} .4269782{col 48}{space 1}    0.77{col 57}{space 3}0.444{col 65}{space 4}  -.51019{col 78}{space 3} 1.163534
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3763615{col 37}{space 2} .5419702{col 48}{space 1}    0.69{col 57}{space 3}0.487{col 65}{space 4}-.6858805{col 78}{space 3} 1.438604
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .7646474{col 37}{space 2} .3589658{col 48}{space 1}    2.13{col 57}{space 3}0.033{col 65}{space 4} .0610873{col 78}{space 3} 1.468207
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est5{txt} stored)

{com}. 
. * Managerial Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(turnover_gs13)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2colreset}{...}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}{col 27} Delta-method
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
turnover_gs13 {c |}{col 15}{res}{space 2} .1139715{col 27}{space 2} .5869462{col 38}{space 1}    0.19{col 47}{space 3}0.846{col 55}{space 4}-1.036422{col 68}{space 3} 1.264365
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Managerial Turnover Rate
. margins, at(turnover_gs13=($turnovergs13_p10 $turnovergs13_p90)) pwcompare(effects)
{res}
{txt}{col 1}Pairwise comparisons of predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.07}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}   Contrast{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} .0079846{col 26}{space 2} .0411529{col 37}{space 1}    0.19{col 46}{space 3}0.846{col 54}{space 4}-.0726736{col 67}{space 3} .0886427
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3D] Unconditional Managerial Turnover Effects on Project Delays
. margins, at(turnover_gs13=(0(0.01)0.15)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:0}}
{lalign 8:2._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.01}}
{lalign 8:3._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.02}}
{lalign 8:4._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.03}}
{lalign 8:5._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.04}}
{lalign 8:6._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.05}}
{lalign 8:7._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.06}}
{lalign 8:8._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.07}}
{lalign 8:9._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.08}}
{lalign 8:10._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.09}}
{lalign 8:11._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.1}}
{lalign 8:12._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.11}}
{lalign 8:13._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.12}}
{lalign 8:14._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.13}}
{lalign 8:15._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.14}}
{lalign 8:16._at: }{space 0}{lalign 13:turnover_gs13} = {res:{ralign 3:.15}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2883923{col 26}{space 2} .0214495{col 37}{space 1}   13.45{col 46}{space 3}0.000{col 54}{space 4} .2463521{col 67}{space 3} .3304326
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .2895264{col 26}{space 2} .0175757{col 37}{space 1}   16.47{col 46}{space 3}0.000{col 54}{space 4} .2550787{col 67}{space 3} .3239742
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .2906628{col 26}{space 2} .0150029{col 37}{space 1}   19.37{col 46}{space 3}0.000{col 54}{space 4} .2612576{col 67}{space 3} .3200679
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2918013{col 26}{space 2} .0144712{col 37}{space 1}   20.16{col 46}{space 3}0.000{col 54}{space 4} .2634382{col 67}{space 3} .3201643
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2929419{col 26}{space 2} .0162073{col 37}{space 1}   18.07{col 46}{space 3}0.000{col 54}{space 4} .2611761{col 67}{space 3} .3247077
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .2940848{col 26}{space 2} .0196382{col 37}{space 1}   14.98{col 46}{space 3}0.000{col 54}{space 4} .2555947{col 67}{space 3} .3325749
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .2952297{col 26}{space 2} .0240652{col 37}{space 1}   12.27{col 46}{space 3}0.000{col 54}{space 4} .2480627{col 67}{space 3} .3423968
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .2963769{col 26}{space 2} .0290498{col 37}{space 1}   10.20{col 46}{space 3}0.000{col 54}{space 4} .2394403{col 67}{space 3} .3533134
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .2975261{col 26}{space 2} .0343612{col 37}{space 1}    8.66{col 46}{space 3}0.000{col 54}{space 4} .2301793{col 67}{space 3}  .364873
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .2986775{col 26}{space 2} .0398788{col 37}{space 1}    7.49{col 46}{space 3}0.000{col 54}{space 4} .2205165{col 67}{space 3} .3768386
{txt}{space 9}11  {c |}{col 14}{res}{space 2}  .299831{col 26}{space 2} .0455361{col 37}{space 1}    6.58{col 46}{space 3}0.000{col 54}{space 4}  .210582{col 67}{space 3} .3890801
{txt}{space 9}12  {c |}{col 14}{res}{space 2} .3009867{col 26}{space 2} .0512942{col 37}{space 1}    5.87{col 46}{space 3}0.000{col 54}{space 4} .2004519{col 67}{space 3} .4015215
{txt}{space 9}13  {c |}{col 14}{res}{space 2} .3021444{col 26}{space 2} .0571294{col 37}{space 1}    5.29{col 46}{space 3}0.000{col 54}{space 4} .1901729{col 67}{space 3} .4141159
{txt}{space 9}14  {c |}{col 14}{res}{space 2} .3033042{col 26}{space 2} .0630262{col 37}{space 1}    4.81{col 46}{space 3}0.000{col 54}{space 4} .1797752{col 67}{space 3} .4268333
{txt}{space 9}15  {c |}{col 14}{res}{space 2} .3044661{col 26}{space 2} .0689742{col 37}{space 1}    4.41{col 46}{space 3}0.000{col 54}{space 4} .1692791{col 67}{space 3} .4396532
{txt}{space 9}16  {c |}{col 14}{res}{space 2} .3056301{col 26}{space 2} .0749664{col 37}{space 1}    4.08{col 46}{space 3}0.000{col 54}{space 4} .1586987{col 67}{space 3} .4525614
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) recastci(rarea) ///
> xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ///
> legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Managerial Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3D. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Managerial Turnover Model [MODEL 5]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram turnover_gs13 if turnover_gs13<=0.15, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:turnover_gs13}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3D.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3D.04-29-2025.gph} saved

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 6, FIGURE 4D] H2: Turnover X Service Length
. 
. eststo: xtgee delay_pct c.turnover_gs13##c.gs13above15_los15pct cio5   activityno_permil_pct cio_tenure lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.turnover_gs13b##c.gs13above15_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.79190332
{txt}Iteration 2:  Tolerance = {res}.14709882
{txt}Iteration 3:  Tolerance = {res}.04154023
{txt}Iteration 4:  Tolerance = {res}.01799563
{txt}Iteration 5:  Tolerance = {res}.01155284
{txt}Iteration 6:  Tolerance = {res}.00840243
{txt}Iteration 7:  Tolerance = {res}.00516292
{txt}Iteration 8:  Tolerance = {res}.00293117
{txt}Iteration 9:  Tolerance = {res}.00159603
{txt}Iteration 10: Tolerance = {res}.00084869
{txt}Iteration 11: Tolerance = {res}.00044497
{txt}Iteration 12: Tolerance = {res}.0002313
{txt}Iteration 13: Tolerance = {res}.00011959
{txt}Iteration 14: Tolerance = {res}.00006163
{txt}Iteration 15: Tolerance = {res}.0000317
{txt}Iteration 16: Tolerance = {res}.00001628
{txt}Iteration 17: Tolerance = {res}8.355e-06
{txt}Iteration 18: Tolerance = {res}4.286e-06
{txt}Iteration 19: Tolerance = {res}2.198e-06
{txt}Iteration 20: Tolerance = {res}1.127e-06
{txt}Iteration 21: Tolerance = {res}5.776e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:29})} = {res}{ralign 6:357.31}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 106:(Std. err. adjusted for clustering on {res:a_id})}
{hline 41}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 42}{c |}{col 54}  Semirobust
{col 1}                               delay_pct{col 42}{c |} Coefficient{col 54}  std. err.{col 66}      z{col 74}   P>|z|{col 82}     [95% con{col 95}f. interval]
{hline 41}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 27}turnover_gs13 {c |}{col 42}{res}{space 2} .1615476{col 54}{space 2} 1.898181{col 65}{space 1}    0.09{col 74}{space 3}0.932{col 82}{space 4}-3.558819{col 95}{space 3} 3.881914
{txt}{space 20}gs13above15_los15pct {c |}{col 42}{res}{space 2}-3.832168{col 54}{space 2}  2.52474{col 65}{space 1}   -1.52{col 74}{space 3}0.129{col 82}{space 4}-8.780568{col 95}{space 3} 1.116232
{txt}{space 40} {c |}
{space 2}c.turnover_gs13#c.gs13above15_los15pct {c |}{col 42}{res}{space 2}  33.5178{col 54}{space 2}  15.0423{col 65}{space 1}    2.23{col 74}{space 3}0.026{col 82}{space 4} 4.035443{col 95}{space 3} 63.00017
{txt}{space 40} {c |}
{space 36}cio5 {c |}{col 42}{res}{space 2}-.1328254{col 54}{space 2} .0827386{col 65}{space 1}   -1.61{col 74}{space 3}0.108{col 82}{space 4}  -.29499{col 95}{space 3} .0293392
{txt}{space 19}activityno_permil_pct {c |}{col 42}{res}{space 2} .3151268{col 54}{space 2} .1973457{col 65}{space 1}    1.60{col 74}{space 3}0.110{col 82}{space 4}-.0716637{col 95}{space 3} .7019172
{txt}{space 30}cio_tenure {c |}{col 42}{res}{space 2}-.0062634{col 54}{space 2} .0239996{col 65}{space 1}   -0.26{col 74}{space 3}0.794{col 82}{space 4}-.0533018{col 95}{space 3}  .040775
{txt}{space 33}lnitemp {c |}{col 42}{res}{space 2} .1782741{col 54}{space 2}  .245885{col 65}{space 1}    0.73{col 74}{space 3}0.468{col 82}{space 4}-.3036517{col 95}{space 3} .6601999
{txt}{space 30}lnprojsize {c |}{col 42}{res}{space 2}   .08354{col 54}{space 2} .0317965{col 65}{space 1}    2.63{col 74}{space 3}0.009{col 82}{space 4}   .02122{col 95}{space 3}   .14586
{txt}{space 29}projects_no {c |}{col 42}{res}{space 2} .0007513{col 54}{space 2} .0018188{col 65}{space 1}    0.41{col 74}{space 3}0.680{col 82}{space 4}-.0028136{col 95}{space 3} .0043161
{txt}{space 33}accrate {c |}{col 42}{res}{space 2} .3562161{col 54}{space 2}  .547793{col 65}{space 1}    0.65{col 74}{space 3}0.516{col 82}{space 4}-.7174384{col 95}{space 3} 1.429871
{txt}{space 25}newprojects_pct {c |}{col 42}{res}{space 2}-.8138825{col 54}{space 2} .1658759{col 65}{space 1}   -4.91{col 74}{space 3}0.000{col 82}{space 4}-1.138993{col 95}{space 3}-.4887716
{txt}{space 38}y1 {c |}{col 42}{res}{space 2}-1.214359{col 54}{space 2} .1070829{col 65}{space 1}  -11.34{col 74}{space 3}0.000{col 82}{space 4}-1.424237{col 95}{space 3} -1.00448
{txt}{space 38}y2 {c |}{col 42}{res}{space 2}-1.179052{col 54}{space 2} .1300482{col 65}{space 1}   -9.07{col 74}{space 3}0.000{col 82}{space 4}-1.433942{col 95}{space 3}-.9241626
{txt}{space 38}y3 {c |}{col 42}{res}{space 2} -1.13431{col 54}{space 2}  .120357{col 65}{space 1}   -9.42{col 74}{space 3}0.000{col 82}{space 4}-1.370205{col 95}{space 3}-.8984144
{txt}{space 38}y4 {c |}{col 42}{res}{space 2}-.8702852{col 54}{space 2} .1180053{col 65}{space 1}   -7.37{col 74}{space 3}0.000{col 82}{space 4}-1.101571{col 95}{space 3}-.6389992
{txt}{space 38}y5 {c |}{col 42}{res}{space 2}-1.001311{col 54}{space 2} .1091107{col 65}{space 1}   -9.18{col 74}{space 3}0.000{col 82}{space 4}-1.215165{col 95}{space 3}-.7874584
{txt}{space 38}y6 {c |}{col 42}{res}{space 2}-.9157721{col 54}{space 2} .1025693{col 65}{space 1}   -8.93{col 74}{space 3}0.000{col 82}{space 4}-1.116804{col 95}{space 3}  -.71474
{txt}{space 38}y7 {c |}{col 42}{res}{space 2}        0{col 54}{txt}  (omitted)
{space 26}turnover_gs13b {c |}{col 42}{res}{space 2} 2.794109{col 54}{space 2} 3.765763{col 65}{space 1}    0.74{col 74}{space 3}0.458{col 82}{space 4} -4.58665{col 95}{space 3} 10.17487
{txt}{space 19}gs13above15_los15pctb {c |}{col 42}{res}{space 2} 10.10597{col 54}{space 2}  6.52744{col 65}{space 1}    1.55{col 74}{space 3}0.122{col 82}{space 4}-2.687579{col 95}{space 3} 22.89952
{txt}{space 40} {c |}
c.turnover_gs13b#c.gs13above15_los15pctb {c |}{col 42}{res}{space 2}-202.7869{col 54}{space 2} 73.39093{col 65}{space 1}   -2.76{col 74}{space 3}0.006{col 82}{space 4}-346.6305{col 95}{space 3}-58.94331
{txt}{space 40} {c |}
{space 18}activityno_permil_pctb {c |}{col 42}{res}{space 2} 1.021245{col 54}{space 2} .3945022{col 65}{space 1}    2.59{col 74}{space 3}0.010{col 82}{space 4} .2480346{col 95}{space 3} 1.794455
{txt}{space 32}lnitempb {c |}{col 42}{res}{space 2}-.1608612{col 54}{space 2} .2480343{col 65}{space 1}   -0.65{col 74}{space 3}0.517{col 82}{space 4}-.6469995{col 95}{space 3} .3252771
{txt}{space 29}lnprojsizeb {c |}{col 42}{res}{space 2}-.0036161{col 54}{space 2} .0657527{col 65}{space 1}   -0.05{col 74}{space 3}0.956{col 82}{space 4} -.132489{col 95}{space 3} .1252567
{txt}{space 28}projects_nob {c |}{col 42}{res}{space 2}-.0048601{col 54}{space 2} .0033875{col 65}{space 1}   -1.43{col 74}{space 3}0.151{col 82}{space 4}-.0114995{col 95}{space 3} .0017792
{txt}{space 32}accrateb {c |}{col 42}{res}{space 2}-1.908487{col 54}{space 2} 1.426308{col 65}{space 1}   -1.34{col 74}{space 3}0.181{col 82}{space 4}   -4.704{col 95}{space 3} .8870255
{txt}{space 24}newprojects_pctb {c |}{col 42}{res}{space 2}-.9447589{col 54}{space 2} .3379359{col 65}{space 1}   -2.80{col 74}{space 3}0.005{col 82}{space 4}-1.607101{col 95}{space 3}-.2824167
{txt}{space 29}cio_tenureb {c |}{col 42}{res}{space 2}-.0392683{col 54}{space 2} .0554389{col 65}{space 1}   -0.71{col 74}{space 3}0.479{col 82}{space 4}-.1479264{col 95}{space 3} .0693899
{txt}{space 18}dmeratio_totalspending {c |}{col 42}{res}{space 2} .1518528{col 54}{space 2} .3750253{col 65}{space 1}    0.40{col 74}{space 3}0.686{col 82}{space 4}-.5831832{col 95}{space 3} .8868888
{txt}{space 17}dmeratio_totalspendingb {c |}{col 42}{res}{space 2} .5436251{col 54}{space 2} .5291789{col 65}{space 1}    1.03{col 74}{space 3}0.304{col 82}{space 4}-.4935464{col 95}{space 3} 1.580797
{txt}{space 35}_cons {c |}{col 42}{res}{space 2} .6031895{col 54}{space 2}  .353602{col 65}{space 1}    1.71{col 74}{space 3}0.088{col 82}{space 4}-.0898576{col 95}{space 3} 1.296237
{txt}{hline 41}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est6{txt} stored)

{com}. 
. * Managerial Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins, dydx(turnover_gs13) at(gs13above15_los15pct=($gs13_los_p10 $gs13_los_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 3:.05}}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}   Contrast{col 27} Delta-method{col 38}    Una{col 47}djusted{col 55}          Una{col 68}djusted
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 10}_at {c |}
{space 6}2 vs 1  {c |}{col 15}{res}{space 2}   .46552{col 27}{space 2} .1829638{col 38}{space 1}    2.54{col 47}{space 3}0.011{col 55}{space 4} .1069174{col 68}{space 3} .8241225
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 4D] Conditional Turnover Effects by Departures' Service Length (H2)
. margins, dydx(turnover_gs13) at(gs13above15_los15pct=(0(0.025)0.2))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:0}}
{lalign 7:2._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.025}}
{lalign 7:3._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.05}}
{lalign 7:4._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.075}}
{lalign 7:5._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.1}}
{lalign 7:6._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.125}}
{lalign 7:7._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.15}}
{lalign 7:8._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.175}}
{lalign 7:9._at: }{space 0}{lalign 16:gs13above15_lo~t} = {res:{ralign 4:.2}}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}{col 27} Delta-method
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 10}_at {c |}
{space 11}1  {c |}{col 15}{res}{space 2} .0488353{col 27}{space 2} .5743001{col 38}{space 1}    0.09{col 47}{space 3}0.932{col 55}{space 4}-1.076772{col 68}{space 3} 1.174443
{txt}{space 11}2  {c |}{col 15}{res}{space 2} .2913561{col 27}{space 2} .5815503{col 38}{space 1}    0.50{col 47}{space 3}0.616{col 55}{space 4}-.8484615{col 68}{space 3} 1.431174
{txt}{space 11}3  {c |}{col 15}{res}{space 2} .5143552{col 27}{space 2} .5895681{col 38}{space 1}    0.87{col 47}{space 3}0.383{col 55}{space 4} -.641177{col 68}{space 3} 1.669888
{txt}{space 11}4  {c |}{col 15}{res}{space 2} .7162928{col 27}{space 2} .5890253{col 38}{space 1}    1.22{col 47}{space 3}0.224{col 55}{space 4}-.4381755{col 68}{space 3} 1.870761
{txt}{space 11}5  {c |}{col 15}{res}{space 2} .8961568{col 27}{space 2} .5740247{col 38}{space 1}    1.56{col 47}{space 3}0.118{col 55}{space 4}-.2289109{col 68}{space 3} 2.021225
{txt}{space 11}6  {c |}{col 15}{res}{space 2} 1.053447{col 27}{space 2} .5418826{col 38}{space 1}    1.94{col 47}{space 3}0.052{col 55}{space 4}-.0086237{col 68}{space 3} 2.115517
{txt}{space 11}7  {c |}{col 15}{res}{space 2} 1.188141{col 27}{space 2} .4927136{col 38}{space 1}    2.41{col 47}{space 3}0.016{col 55}{space 4} .2224404{col 68}{space 3} 2.153842
{txt}{space 11}8  {c |}{col 15}{res}{space 2} 1.300657{col 27}{space 2} .4293634{col 38}{space 1}    3.03{col 47}{space 3}0.002{col 55}{space 4} .4591198{col 68}{space 3} 2.142193
{txt}{space 11}9  {c |}{col 15}{res}{space 2} 1.391791{col 27}{space 2} .3582268{col 38}{space 1}    3.89{col 47}{space 3}0.000{col 55}{space 4} .6896794{col 68}{space 3} 2.093903
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> xlabel(0 "0" 0.1 "10" 0.2 "20" ,format(%9.0f) ) ylabel( ,format(%9.1f) angle(0)) ///
> legend(off) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departing Managers' Service Length (>15 years, %)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 4D. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Managerial Turnover Model [MODEL 6]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram gs13above15_los15pct if gs13above15_los15pct<=0.2 , fc(gray%20) lc(gray%10) yaxis(2) yscale(axis(2) alt) xlabel(0 "0" 0.1 "10" 0.2 "20" ,format(%9.0f)) ylabel( ,format(%9.1f) angle(0) ) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:gs13above15_los15pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4D.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4D.04-29-2025.gph} saved

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 7, FIGURE 5D] H3: Turnover X Agency Centralization
. 
. eststo: xtgee delay_pct c.turnover_gs13##i.cio5 c.activityno_permil_pct cio_tenure newprojects_pct gs13above15_los15pct lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.turnover_gs13b c.activityno_permil_pctb gs13above15_los15pctb  lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:newprojects_pct} omitted because of collinearity.
note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}1.2385937
{txt}Iteration 2:  Tolerance = {res}.19472637
{txt}Iteration 3:  Tolerance = {res}.05899281
{txt}Iteration 4:  Tolerance = {res}.02143025
{txt}Iteration 5:  Tolerance = {res}.00992351
{txt}Iteration 6:  Tolerance = {res}.00457067
{txt}Iteration 7:  Tolerance = {res}.00203634
{txt}Iteration 8:  Tolerance = {res}.00089366
{txt}Iteration 9:  Tolerance = {res}.00038949
{txt}Iteration 10: Tolerance = {res}.00016923
{txt}Iteration 11: Tolerance = {res}.00007344
{txt}Iteration 12: Tolerance = {res}.00003186
{txt}Iteration 13: Tolerance = {res}.00001383
{txt}Iteration 14: Tolerance = {res}6.000e-06
{txt}Iteration 15: Tolerance = {res}2.605e-06
{txt}Iteration 16: Tolerance = {res}1.131e-06
{txt}Iteration 17: Tolerance = {res}4.915e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:323.45}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 10}turnover_gs13 {c |}{col 25}{res}{space 2} .4361928{col 37}{space 2} 1.921185{col 48}{space 1}    0.23{col 57}{space 3}0.820{col 65}{space 4} -3.32926{col 78}{space 3} 4.201646
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.0362835{col 37}{space 2} .1166823{col 48}{space 1}   -0.31{col 57}{space 3}0.756{col 65}{space 4}-.2649766{col 78}{space 3} .1924095
{txt}{space 23} {c |}
{space 3}cio5#c.turnover_gs13 {c |}
{space 21}1  {c |}{col 25}{res}{space 2}-1.930724{col 37}{space 2}  3.10557{col 48}{space 1}   -0.62{col 57}{space 3}0.534{col 65}{space 4}-8.017529{col 78}{space 3}  4.15608
{txt}{space 23} {c |}
{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2841718{col 37}{space 2} .1970461{col 48}{space 1}    1.44{col 57}{space 3}0.149{col 65}{space 4}-.1020314{col 78}{space 3} .6703751
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0145706{col 37}{space 2} .0242678{col 48}{space 1}   -0.60{col 57}{space 3}0.548{col 65}{space 4}-.0621347{col 78}{space 3} .0329935
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8065545{col 37}{space 2} .1654672{col 48}{space 1}   -4.87{col 57}{space 3}0.000{col 65}{space 4}-1.130864{col 78}{space 3}-.4822447
{txt}{space 3}gs13above15_los15pct {c |}{col 25}{res}{space 2}-.1460687{col 37}{space 2} 1.042899{col 48}{space 1}   -0.14{col 57}{space 3}0.889{col 65}{space 4}-2.190113{col 78}{space 3} 1.897976
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .1187452{col 37}{space 2}  .238903{col 48}{space 1}    0.50{col 57}{space 3}0.619{col 65}{space 4} -.349496{col 78}{space 3} .5869864
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0810084{col 37}{space 2} .0271307{col 48}{space 1}    2.99{col 57}{space 3}0.003{col 65}{space 4} .0278331{col 78}{space 3} .1341836
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0009892{col 37}{space 2} .0016604{col 48}{space 1}    0.60{col 57}{space 3}0.551{col 65}{space 4}-.0022651{col 78}{space 3} .0042435
{txt}{space 16}accrate {c |}{col 25}{res}{space 2}  .332889{col 37}{space 2} .5429045{col 48}{space 1}    0.61{col 57}{space 3}0.540{col 65}{space 4}-.7311843{col 78}{space 3} 1.396962
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 21}y1 {c |}{col 25}{res}{space 2}-1.195349{col 37}{space 2} .1045353{col 48}{space 1}  -11.43{col 57}{space 3}0.000{col 65}{space 4}-1.400235{col 78}{space 3} -.990464
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.178575{col 37}{space 2} .1267659{col 48}{space 1}   -9.30{col 57}{space 3}0.000{col 65}{space 4}-1.427032{col 78}{space 3}-.9301183
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.121108{col 37}{space 2} .1183271{col 48}{space 1}   -9.47{col 57}{space 3}0.000{col 65}{space 4}-1.353025{col 78}{space 3}-.8891915
{txt}{space 21}y4 {c |}{col 25}{res}{space 2} -.878361{col 37}{space 2} .1221957{col 48}{space 1}   -7.19{col 57}{space 3}0.000{col 65}{space 4} -1.11786{col 78}{space 3}-.6388617
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9965838{col 37}{space 2} .1096941{col 48}{space 1}   -9.09{col 57}{space 3}0.000{col 65}{space 4} -1.21158{col 78}{space 3}-.7815874
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8973829{col 37}{space 2} .1023634{col 48}{space 1}   -8.77{col 57}{space 3}0.000{col 65}{space 4}-1.098011{col 78}{space 3}-.6967544
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 9}turnover_gs13b {c |}{col 25}{res}{space 2}  .189048{col 37}{space 2}  3.30502{col 48}{space 1}    0.06{col 57}{space 3}0.954{col 65}{space 4}-6.288672{col 78}{space 3} 6.666768
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9931109{col 37}{space 2} .4091932{col 48}{space 1}    2.43{col 57}{space 3}0.015{col 65}{space 4} .1911069{col 78}{space 3} 1.795115
{txt}{space 2}gs13above15_los15pctb {c |}{col 25}{res}{space 2}-6.604651{col 37}{space 2} 2.301324{col 48}{space 1}   -2.87{col 57}{space 3}0.004{col 65}{space 4}-11.11516{col 78}{space 3}-2.094138
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2}-.1107491{col 37}{space 2}   .24424{col 48}{space 1}   -0.45{col 57}{space 3}0.650{col 65}{space 4}-.5894507{col 78}{space 3} .3679524
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0009105{col 37}{space 2} .0635992{col 48}{space 1}    0.01{col 57}{space 3}0.989{col 65}{space 4}-.1237416{col 78}{space 3} .1255626
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0055676{col 37}{space 2} .0032668{col 48}{space 1}   -1.70{col 57}{space 3}0.088{col 65}{space 4}-.0119703{col 78}{space 3} .0008352
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-.9030211{col 37}{space 2} 1.235712{col 48}{space 1}   -0.73{col 57}{space 3}0.465{col 65}{space 4}-3.324972{col 78}{space 3} 1.518929
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.013575{col 37}{space 2} .3394823{col 48}{space 1}   -2.99{col 57}{space 3}0.003{col 65}{space 4}-1.678948{col 78}{space 3}-.3482015
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2} -.032005{col 37}{space 2} .0564738{col 48}{space 1}   -0.57{col 57}{space 3}0.571{col 65}{space 4}-.1426917{col 78}{space 3} .0786817
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .3169768{col 37}{space 2} .4281193{col 48}{space 1}    0.74{col 57}{space 3}0.459{col 65}{space 4}-.5221216{col 78}{space 3} 1.156075
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .4125234{col 37}{space 2} .5540887{col 48}{space 1}    0.74{col 57}{space 3}0.457{col 65}{space 4}-.6734706{col 78}{space 3} 1.498517
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .7484384{col 37}{space 2} .3680538{col 48}{space 1}    2.03{col 57}{space 3}0.042{col 65}{space 4} .0270661{col 78}{space 3} 1.469811
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est7{txt} stored)

{com}. 
. * Mangerial Turnover Rate: Marginal Effect Differentials Based on Agency Centralization
. margins , dydx(turnover_gs13) over(cio5) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}   Contrast{col 27} Delta-method{col 38}    Una{col 47}djusted{col 55}          Una{col 68}djusted
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 9}cio5 {c |}
{space 6}1 vs 0  {c |}{col 15}{res}{space 2}-.5621324{col 27}{space 2} .8961787{col 38}{space 1}   -0.63{col 47}{space 3}0.530{col 55}{space 4} -2.31861{col 68}{space 3} 1.194346
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 5D] Conditional Turnover Effects by Agency Centralization (H3)
. margins , dydx(turnover_gs13) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}{col 27} Delta-method
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 9}cio5 {c |}
{space 11}0  {c |}{col 15}{res}{space 2} .1317017{col 27}{space 2} .5805355{col 38}{space 1}    0.23{col 47}{space 3}0.821{col 55}{space 4}-1.006127{col 68}{space 3}  1.26953
{txt}{space 11}1  {c |}{col 15}{res}{space 2}-.4304307{col 27}{space 2}  1.02388{col 38}{space 1}   -0.42{col 47}{space 3}0.674{col 55}{space 4}-2.437199{col 68}{space 3} 1.576338
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(black)) ciopt(color(black%60))  yline(0, lcolor(red) lpattern(shortdash)) ///
> xlabel (-1 " " -0.1 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5D. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Managerial Turnover Model [MODEL 7]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5D.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5D.04-29-2025.gph} saved

{com}. 
. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 8, FIGURE 6D] H4: Turnover X Task Standardization
. 
. eststo: xtgee delay_pct c.turnover_gs13##c.activityno_permil_pct newprojects_pct gs13above15_los15pct cio5 cio_tenure lnitemp lnprojsize projects_no accrate  y1-y7 c.turnover_gs13b c.activityno_permil_pctb gs13above15_los15pctb  lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.32390252
{txt}Iteration 2:  Tolerance = {res}.08788987
{txt}Iteration 3:  Tolerance = {res}.03334301
{txt}Iteration 4:  Tolerance = {res}.01296014
{txt}Iteration 5:  Tolerance = {res}.00688835
{txt}Iteration 6:  Tolerance = {res}.00331653
{txt}Iteration 7:  Tolerance = {res}.00152504
{txt}Iteration 8:  Tolerance = {res}.00068494
{txt}Iteration 9:  Tolerance = {res}.00030389
{txt}Iteration 10: Tolerance = {res}.00013401
{txt}Iteration 11: Tolerance = {res}.00005893
{txt}Iteration 12: Tolerance = {res}.00002589
{txt}Iteration 13: Tolerance = {res}.00001138
{txt}Iteration 14: Tolerance = {res}4.999e-06
{txt}Iteration 15: Tolerance = {res}2.198e-06
{txt}Iteration 16: Tolerance = {res}9.672e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:334.60}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 105:(Std. err. adjusted for clustering on {res:a_id})}
{hline 40}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 41}{c |}{col 53}  Semirobust
{col 1}                              delay_pct{col 41}{c |} Coefficient{col 53}  std. err.{col 65}      z{col 73}   P>|z|{col 81}     [95% con{col 94}f. interval]
{hline 40}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}turnover_gs13 {c |}{col 41}{res}{space 2}  2.51362{col 53}{space 2} 2.811491{col 64}{space 1}    0.89{col 73}{space 3}0.371{col 81}{space 4}-2.996801{col 94}{space 3} 8.024041
{txt}{space 18}activityno_permil_pct {c |}{col 41}{res}{space 2} .4073762{col 53}{space 2} .2303701{col 64}{space 1}    1.77{col 73}{space 3}0.077{col 81}{space 4} -.044141{col 94}{space 3} .8588934
{txt}{space 39} {c |}
c.turnover_gs13#c.activityno_permil_pct {c |}{col 41}{res}{space 2}-5.219138{col 53}{space 2} 3.836477{col 64}{space 1}   -1.36{col 73}{space 3}0.174{col 81}{space 4} -12.7385{col 94}{space 3}  2.30022
{txt}{space 39} {c |}
{space 24}newprojects_pct {c |}{col 41}{res}{space 2}-.8259931{col 53}{space 2} .1653531{col 64}{space 1}   -5.00{col 73}{space 3}0.000{col 81}{space 4}-1.150079{col 94}{space 3}-.5019071
{txt}{space 19}gs13above15_los15pct {c |}{col 41}{res}{space 2}-.6730199{col 53}{space 2} 1.320628{col 64}{space 1}   -0.51{col 73}{space 3}0.610{col 81}{space 4}-3.261403{col 94}{space 3} 1.915363
{txt}{space 35}cio5 {c |}{col 41}{res}{space 2}-.0927504{col 53}{space 2} .0828527{col 64}{space 1}   -1.12{col 73}{space 3}0.263{col 81}{space 4}-.2551386{col 94}{space 3} .0696378
{txt}{space 29}cio_tenure {c |}{col 41}{res}{space 2}-.0095015{col 53}{space 2} .0242666{col 64}{space 1}   -0.39{col 73}{space 3}0.695{col 81}{space 4}-.0570633{col 94}{space 3} .0380602
{txt}{space 32}lnitemp {c |}{col 41}{res}{space 2} .0634786{col 53}{space 2} .2208732{col 64}{space 1}    0.29{col 73}{space 3}0.774{col 81}{space 4} -.369425{col 94}{space 3} .4963821
{txt}{space 29}lnprojsize {c |}{col 41}{res}{space 2}  .079746{col 53}{space 2} .0299727{col 64}{space 1}    2.66{col 73}{space 3}0.008{col 81}{space 4} .0210005{col 94}{space 3} .1384915
{txt}{space 28}projects_no {c |}{col 41}{res}{space 2} .0012494{col 53}{space 2} .0017469{col 64}{space 1}    0.72{col 73}{space 3}0.474{col 81}{space 4}-.0021744{col 94}{space 3} .0046732
{txt}{space 32}accrate {c |}{col 41}{res}{space 2} .4048238{col 53}{space 2} .5436321{col 64}{space 1}    0.74{col 73}{space 3}0.456{col 81}{space 4}-.6606756{col 94}{space 3} 1.470323
{txt}{space 37}y1 {c |}{col 41}{res}{space 2}-1.212583{col 53}{space 2} .1041732{col 64}{space 1}  -11.64{col 73}{space 3}0.000{col 81}{space 4}-1.416758{col 94}{space 3}-1.008407
{txt}{space 37}y2 {c |}{col 41}{res}{space 2} -1.19637{col 53}{space 2} .1262724{col 64}{space 1}   -9.47{col 73}{space 3}0.000{col 81}{space 4}-1.443859{col 94}{space 3}-.9488804
{txt}{space 37}y3 {c |}{col 41}{res}{space 2}-1.133248{col 53}{space 2} .1179233{col 64}{space 1}   -9.61{col 73}{space 3}0.000{col 81}{space 4}-1.364374{col 94}{space 3}-.9021228
{txt}{space 37}y4 {c |}{col 41}{res}{space 2}-.8779104{col 53}{space 2} .1199075{col 64}{space 1}   -7.32{col 73}{space 3}0.000{col 81}{space 4}-1.112925{col 94}{space 3} -.642896
{txt}{space 37}y5 {c |}{col 41}{res}{space 2}-1.011468{col 53}{space 2} .1097669{col 64}{space 1}   -9.21{col 73}{space 3}0.000{col 81}{space 4}-1.226607{col 94}{space 3}-.7963293
{txt}{space 37}y6 {c |}{col 41}{res}{space 2}-.9079056{col 53}{space 2} .1021234{col 64}{space 1}   -8.89{col 73}{space 3}0.000{col 81}{space 4}-1.108064{col 94}{space 3}-.7077474
{txt}{space 37}y7 {c |}{col 41}{res}{space 2}        0{col 53}{txt}  (omitted)
{space 25}turnover_gs13b {c |}{col 41}{res}{space 2}-.1044884{col 53}{space 2} 3.357388{col 64}{space 1}   -0.03{col 73}{space 3}0.975{col 81}{space 4}-6.684847{col 94}{space 3} 6.475871
{txt}{space 17}activityno_permil_pctb {c |}{col 41}{res}{space 2} 1.024688{col 53}{space 2} .3996831{col 64}{space 1}    2.56{col 73}{space 3}0.010{col 81}{space 4} .2413233{col 94}{space 3} 1.808052
{txt}{space 18}gs13above15_los15pctb {c |}{col 41}{res}{space 2}-6.469487{col 53}{space 2} 2.227174{col 64}{space 1}   -2.90{col 73}{space 3}0.004{col 81}{space 4}-10.83467{col 94}{space 3}-2.104305
{txt}{space 31}lnitempb {c |}{col 41}{res}{space 2}-.0543888{col 53}{space 2} .2273567{col 64}{space 1}   -0.24{col 73}{space 3}0.811{col 81}{space 4}-.4999998{col 94}{space 3} .3912222
{txt}{space 28}lnprojsizeb {c |}{col 41}{res}{space 2}  .004255{col 53}{space 2} .0632796{col 64}{space 1}    0.07{col 73}{space 3}0.946{col 81}{space 4}-.1197708{col 94}{space 3} .1282808
{txt}{space 27}projects_nob {c |}{col 41}{res}{space 2}-.0056689{col 53}{space 2} .0032862{col 64}{space 1}   -1.73{col 73}{space 3}0.085{col 81}{space 4}-.0121098{col 94}{space 3} .0007719
{txt}{space 31}accrateb {c |}{col 41}{res}{space 2}-1.028698{col 53}{space 2} 1.235251{col 64}{space 1}   -0.83{col 73}{space 3}0.405{col 81}{space 4}-3.449746{col 94}{space 3}  1.39235
{txt}{space 23}newprojects_pctb {c |}{col 41}{res}{space 2}-.9906926{col 53}{space 2} .3349418{col 64}{space 1}   -2.96{col 73}{space 3}0.003{col 81}{space 4}-1.647166{col 94}{space 3}-.3342188
{txt}{space 28}cio_tenureb {c |}{col 41}{res}{space 2}-.0361939{col 53}{space 2} .0558189{col 64}{space 1}   -0.65{col 73}{space 3}0.517{col 81}{space 4} -.145597{col 94}{space 3} .0732092
{txt}{space 17}dmeratio_totalspending {c |}{col 41}{res}{space 2}  .311084{col 53}{space 2} .4130787{col 64}{space 1}    0.75{col 73}{space 3}0.451{col 81}{space 4}-.4985354{col 94}{space 3} 1.120703
{txt}{space 16}dmeratio_totalspendingb {c |}{col 41}{res}{space 2} .3810046{col 53}{space 2} .5350095{col 64}{space 1}    0.71{col 73}{space 3}0.476{col 81}{space 4}-.6675948{col 94}{space 3} 1.429604
{txt}{space 34}_cons {c |}{col 41}{res}{space 2} .7068892{col 53}{space 2} .3607223{col 64}{space 1}    1.96{col 73}{space 3}0.050{col 81}{space 4}-.0001134{col 94}{space 3} 1.413892
{txt}{hline 40}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est8{txt} stored)

{com}. 
. * Managerial Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Task Standardization
. margins , dydx(turnover_gs13) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}   Contrast{col 27} Delta-method{col 38}    Una{col 47}djusted{col 55}          Una{col 68}djusted
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 10}_at {c |}
{space 6}2 vs 1  {c |}{col 15}{res}{space 2}-1.098379{col 27}{space 2} .7999964{col 38}{space 1}   -1.37{col 47}{space 3}0.170{col 55}{space 4}-2.666343{col 68}{space 3} .4695855
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 6D] Conditional Turnover Effects by Task Standardization (H4)
. margins , dydx(turnover_gs13) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:turnover_gs13}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}{col 27} Delta-method
{col 15}{c |}      dy/dx{col 27}   std. err.{col 39}      z{col 47}   P>|z|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}turnover_gs13 {txt}{c |}
{space 10}_at {c |}
{space 11}1  {c |}{col 15}{res}{space 2} .7242871{col 27}{space 2}  .810036{col 38}{space 1}    0.89{col 47}{space 3}0.371{col 55}{space 4}-.8633542{col 68}{space 3} 2.311928
{txt}{space 11}2  {c |}{col 15}{res}{space 2} .5819536{col 27}{space 2} .7418698{col 38}{space 1}    0.78{col 47}{space 3}0.433{col 55}{space 4}-.8720845{col 68}{space 3} 2.035992
{txt}{space 11}3  {c |}{col 15}{res}{space 2} .4352177{col 27}{space 2} .6804243{col 38}{space 1}    0.64{col 47}{space 3}0.522{col 55}{space 4}-.8983895{col 68}{space 3} 1.768825
{txt}{space 11}4  {c |}{col 15}{res}{space 2} .2842643{col 27}{space 2} .6293882{col 38}{space 1}    0.45{col 47}{space 3}0.652{col 55}{space 4}-.9493138{col 68}{space 3} 1.517842
{txt}{space 11}5  {c |}{col 15}{res}{space 2} .1292975{col 27}{space 2} .5932582{col 38}{space 1}    0.22{col 47}{space 3}0.827{col 55}{space 4}-1.033467{col 68}{space 3} 1.292062
{txt}{space 11}6  {c |}{col 15}{res}{space 2}-.0294598{col 27}{space 2} .5766343{col 38}{space 1}   -0.05{col 47}{space 3}0.959{col 55}{space 4}-1.159642{col 68}{space 3} 1.100723
{txt}{space 11}7  {c |}{col 15}{res}{space 2}-.1917661{col 27}{space 2} .5828437{col 38}{space 1}   -0.33{col 47}{space 3}0.742{col 55}{space 4}-1.334119{col 68}{space 3} .9505865
{txt}{space 11}8  {c |}{col 15}{res}{space 2}-.3573617{col 27}{space 2} .6126086{col 38}{space 1}   -0.58{col 47}{space 3}0.560{col 55}{space 4}-1.558052{col 68}{space 3}  .843329
{txt}{space 11}9  {c |}{col 15}{res}{space 2}-.5259696{col 27}{space 2} .6638867{col 38}{space 1}   -0.79{col 47}{space 3}0.428{col 55}{space 4}-1.827164{col 68}{space 3} .7752244
{txt}{space 10}10  {c |}{col 15}{res}{space 2}-.6972959{col 27}{space 2} .7329857{col 38}{space 1}   -0.95{col 47}{space 3}0.341{col 55}{space 4}-2.133922{col 68}{space 3} .7393297
{txt}{space 10}11  {c |}{col 15}{res}{space 2}-.8710311{col 27}{space 2} .8159045{col 38}{space 1}   -1.07{col 47}{space 3}0.286{col 55}{space 4}-2.470175{col 68}{space 3} .7281124
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", margin(t=1) size(11pt)) ///
> title("{c -(}bf: Figure 6D. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Managerial Turnover Model [MODEL 8]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel( ,format(%9.1f) angle(0))  ylabel( ,format(%9.1f) angle(0) axis(2)) ytitle(, axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6D.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6D.04-29-2025.gph} saved

{com}. *
. ****************************************************************************************************************************************************
. * [TABLE A1. MODELS A1 - A4; AGGREGATE TURNOVER MODELS]
. cd "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES"
{res}C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES
{txt}
{com}. 
. esttab est1 est2 est3 est4 using ManuscriptTable1.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 1. Fractional Panel Probit Analysis of the Impact of Aggregate Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 1" "Model 2" "Model 3" "Model 4" )  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) replace
{res}{txt}(output written to {browse  `"ManuscriptTable1.04-29-2025.rtf"'})

{com}. 
. esttab est5 est6 est7 est8 using ManuscriptTable1.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 1. Fractional Panel Probit Analysis of the Impact of Aggregate Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 5" "Model 6" "Model 7" "Model 8")  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) append
{res}{txt}(output written to {browse  `"ManuscriptTable1.04-29-2025.rtf"'})

{com}. ****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. discard
{txt}
{com}. 
. **# * [MODEL 9 - 12] [FIGURES 3B-6B] AGGREGATE VOLUNTARY TURNOVER MODELS
. 
. 
. * [MODEL 9, FIGURE 3B] H1: Aggregate Voluntary Turnover Baseline Effects
. eststo: xtgee delay_pct c.voluntary volabove15_los15pct cio5 activityno_permil_pct lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 voluntaryb volabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.63237996
{txt}Iteration 2:  Tolerance = {res}.10895436
{txt}Iteration 3:  Tolerance = {res}.03111716
{txt}Iteration 4:  Tolerance = {res}.00748958
{txt}Iteration 5:  Tolerance = {res}.00367466
{txt}Iteration 6:  Tolerance = {res}.00209307
{txt}Iteration 7:  Tolerance = {res}.00108083
{txt}Iteration 8:  Tolerance = {res}.00053044
{txt}Iteration 9:  Tolerance = {res}.00025259
{txt}Iteration 10: Tolerance = {res}.00011784
{txt}Iteration 11: Tolerance = {res}.00005418
{txt}Iteration 12: Tolerance = {res}.00002464
{txt}Iteration 13: Tolerance = {res}.00001112
{txt}Iteration 14: Tolerance = {res}4.986e-06
{txt}Iteration 15: Tolerance = {res}2.226e-06
{txt}Iteration 16: Tolerance = {res}9.902e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:375.60}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 14}voluntary {c |}{col 25}{res}{space 2} 2.254866{col 37}{space 2} 1.319756{col 48}{space 1}    1.71{col 57}{space 3}0.088{col 65}{space 4}-.3318086{col 78}{space 3}  4.84154
{txt}{space 4}volabove15_los15pct {c |}{col 25}{res}{space 2}-.2137376{col 37}{space 2} 1.032992{col 48}{space 1}   -0.21{col 57}{space 3}0.836{col 65}{space 4}-2.238364{col 78}{space 3} 1.810889
{txt}{space 19}cio5 {c |}{col 25}{res}{space 2}  -.02558{col 37}{space 2} .0937177{col 48}{space 1}   -0.27{col 57}{space 3}0.785{col 65}{space 4}-.2092633{col 78}{space 3} .1581033
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2245808{col 37}{space 2} .2318008{col 48}{space 1}    0.97{col 57}{space 3}0.333{col 65}{space 4}-.2297404{col 78}{space 3} .6789021
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.3877922{col 37}{space 2} .2789893{col 48}{space 1}   -1.39{col 57}{space 3}0.165{col 65}{space 4}-.9346013{col 78}{space 3} .1590168
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0391424{col 37}{space 2}  .029129{col 48}{space 1}    1.34{col 57}{space 3}0.179{col 65}{space 4}-.0179493{col 78}{space 3} .0962342
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0012308{col 37}{space 2} .0016401{col 48}{space 1}    0.75{col 57}{space 3}0.453{col 65}{space 4}-.0019837{col 78}{space 3} .0044452
{txt}{space 16}accrate {c |}{col 25}{res}{space 2}-.0205327{col 37}{space 2}  .549539{col 48}{space 1}   -0.04{col 57}{space 3}0.970{col 65}{space 4}-1.097609{col 78}{space 3} 1.056544
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8615316{col 37}{space 2} .1659728{col 48}{space 1}   -5.19{col 57}{space 3}0.000{col 65}{space 4}-1.186832{col 78}{space 3}-.5362308
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0132352{col 37}{space 2} .0270582{col 48}{space 1}   -0.49{col 57}{space 3}0.625{col 65}{space 4}-.0662683{col 78}{space 3} .0397978
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.288434{col 37}{space 2}  .099293{col 48}{space 1}  -12.98{col 57}{space 3}0.000{col 65}{space 4}-1.483044{col 78}{space 3}-1.093823
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.219873{col 37}{space 2} .1220109{col 48}{space 1}  -10.00{col 57}{space 3}0.000{col 65}{space 4} -1.45901{col 78}{space 3}-.9807362
{txt}{space 21}y3 {c |}{col 25}{res}{space 2} -1.18653{col 37}{space 2} .1209881{col 48}{space 1}   -9.81{col 57}{space 3}0.000{col 65}{space 4}-1.423663{col 78}{space 3}-.9493979
{txt}{space 21}y4 {c |}{col 25}{res}{space 2} -.976817{col 37}{space 2} .1274337{col 48}{space 1}   -7.67{col 57}{space 3}0.000{col 65}{space 4}-1.226583{col 78}{space 3}-.7270515
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-1.048901{col 37}{space 2}  .110612{col 48}{space 1}   -9.48{col 57}{space 3}0.000{col 65}{space 4}-1.265697{col 78}{space 3}-.8321059
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-1.023742{col 37}{space 2} .1121815{col 48}{space 1}   -9.13{col 57}{space 3}0.000{col 65}{space 4}-1.243613{col 78}{space 3}-.8038699
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 13}voluntaryb {c |}{col 25}{res}{space 2}-.2569336{col 37}{space 2}  2.78184{col 48}{space 1}   -0.09{col 57}{space 3}0.926{col 65}{space 4}-5.709241{col 78}{space 3} 5.195374
{txt}{space 3}volabove15_los15pctb {c |}{col 25}{res}{space 2}-7.575675{col 37}{space 2} 2.320595{col 48}{space 1}   -3.26{col 57}{space 3}0.001{col 65}{space 4}-12.12396{col 78}{space 3}-3.027392
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9137856{col 37}{space 2} .4130828{col 48}{space 1}    2.21{col 57}{space 3}0.027{col 65}{space 4} .1041583{col 78}{space 3} 1.723413
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .3807869{col 37}{space 2} .2846967{col 48}{space 1}    1.34{col 57}{space 3}0.181{col 65}{space 4}-.1772083{col 78}{space 3} .9387821
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0353171{col 37}{space 2}   .06119{col 48}{space 1}    0.58{col 57}{space 3}0.564{col 65}{space 4}-.0846132{col 78}{space 3} .1552474
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2} -.005338{col 37}{space 2} .0029219{col 48}{space 1}   -1.83{col 57}{space 3}0.068{col 65}{space 4}-.0110649{col 78}{space 3} .0003889
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-1.044856{col 37}{space 2} 1.256236{col 48}{space 1}   -0.83{col 57}{space 3}0.406{col 65}{space 4}-3.507032{col 78}{space 3} 1.417321
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.103126{col 37}{space 2} .3344248{col 48}{space 1}   -3.30{col 57}{space 3}0.001{col 65}{space 4}-1.758587{col 78}{space 3}-.4476657
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0384576{col 37}{space 2} .0612063{col 48}{space 1}   -0.63{col 57}{space 3}0.530{col 65}{space 4}-.1584198{col 78}{space 3} .0815046
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .0518569{col 37}{space 2} .3879536{col 48}{space 1}    0.13{col 57}{space 3}0.894{col 65}{space 4}-.7085183{col 78}{space 3}  .812232
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3996639{col 37}{space 2}  .468999{col 48}{space 1}    0.85{col 57}{space 3}0.394{col 65}{space 4}-.5195572{col 78}{space 3} 1.318885
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} 1.112997{col 37}{space 2} .3099703{col 48}{space 1}    3.59{col 57}{space 3}0.000{col 65}{space 4}  .505466{col 78}{space 3} 1.720527
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est1{txt} stored)

{com}. 
. * Aggregate Voluntary Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(voluntary)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 3}voluntary {c |}{col 14}{res}{space 2} .6724829{col 26}{space 2} .3958707{col 37}{space 1}    1.70{col 46}{space 3}0.089{col 54}{space 4}-.1034094{col 67}{space 3} 1.448375
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Aggregate Voluntary Turnover Rate
. margins, at(voluntary=($voluntary_p10 $voluntary_p90)) pwcompare(effects)
{res}
{txt}{col 1}Pairwise comparisons of predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.08}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}   Contrast{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} .0536608{col 26}{space 2} .0314696{col 37}{space 1}    1.71{col 46}{space 3}0.088{col 54}{space 4}-.0080185{col 67}{space 3} .1153401
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3B] Unconditional Turnover Effects on Project Delays (H1)
. margins, at(voluntary=(0(0.01)0.15)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:0}}
{lalign 8:2._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.01}}
{lalign 8:3._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.02}}
{lalign 8:4._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.03}}
{lalign 8:5._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.04}}
{lalign 8:6._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.05}}
{lalign 8:7._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.06}}
{lalign 8:8._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.07}}
{lalign 8:9._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.08}}
{lalign 8:10._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.09}}
{lalign 8:11._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.1}}
{lalign 8:12._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.11}}
{lalign 8:13._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.12}}
{lalign 8:14._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.13}}
{lalign 8:15._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.14}}
{lalign 8:16._at: }{space 0}{lalign 9:voluntary} = {res:{ralign 3:.15}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2682076{col 26}{space 2} .0200949{col 37}{space 1}   13.35{col 46}{space 3}0.000{col 54}{space 4} .2288223{col 67}{space 3} .3075929
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .2746484{col 26}{space 2} .0177119{col 37}{space 1}   15.51{col 46}{space 3}0.000{col 54}{space 4} .2399337{col 67}{space 3} .3093632
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .2811687{col 26}{space 2} .0157568{col 37}{space 1}   17.84{col 46}{space 3}0.000{col 54}{space 4}  .250286{col 67}{space 3} .3120514
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2877669{col 26}{space 2} .0144787{col 37}{space 1}   19.88{col 46}{space 3}0.000{col 54}{space 4} .2593891{col 67}{space 3} .3161447
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2944414{col 26}{space 2} .0141386{col 37}{space 1}   20.83{col 46}{space 3}0.000{col 54}{space 4} .2667303{col 67}{space 3} .3221525
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .3011906{col 26}{space 2} .0148711{col 37}{space 1}   20.25{col 46}{space 3}0.000{col 54}{space 4} .2720437{col 67}{space 3} .3303375
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .3080127{col 26}{space 2} .0165956{col 37}{space 1}   18.56{col 46}{space 3}0.000{col 54}{space 4}  .275486{col 67}{space 3} .3405395
{txt}{space 10}8  {c |}{col 14}{res}{space 2}  .314906{col 26}{space 2}  .019096{col 37}{space 1}   16.49{col 46}{space 3}0.000{col 54}{space 4} .2774786{col 67}{space 3} .3523334
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .3218684{col 26}{space 2} .0221533{col 37}{space 1}   14.53{col 46}{space 3}0.000{col 54}{space 4} .2784489{col 67}{space 3}  .365288
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .3288981{col 26}{space 2} .0256037{col 37}{space 1}   12.85{col 46}{space 3}0.000{col 54}{space 4} .2787158{col 67}{space 3} .3790805
{txt}{space 9}11  {c |}{col 14}{res}{space 2}  .335993{col 26}{space 2} .0293378{col 37}{space 1}   11.45{col 46}{space 3}0.000{col 54}{space 4} .2784919{col 67}{space 3} .3934941
{txt}{space 9}12  {c |}{col 14}{res}{space 2} .3431508{col 26}{space 2} .0332841{col 37}{space 1}   10.31{col 46}{space 3}0.000{col 54}{space 4} .2779152{col 67}{space 3} .4083865
{txt}{space 9}13  {c |}{col 14}{res}{space 2} .3503695{col 26}{space 2} .0373949{col 37}{space 1}    9.37{col 46}{space 3}0.000{col 54}{space 4} .2770769{col 67}{space 3} .4236622
{txt}{space 9}14  {c |}{col 14}{res}{space 2} .3576468{col 26}{space 2} .0416375{col 37}{space 1}    8.59{col 46}{space 3}0.000{col 54}{space 4} .2760388{col 67}{space 3} .4392547
{txt}{space 9}15  {c |}{col 14}{res}{space 2} .3649802{col 26}{space 2} .0459881{col 37}{space 1}    7.94{col 46}{space 3}0.000{col 54}{space 4} .2748452{col 67}{space 3} .4551152
{txt}{space 9}16  {c |}{col 14}{res}{space 2} .3723674{col 26}{space 2} .0504288{col 37}{space 1}    7.38{col 46}{space 3}0.000{col 54}{space 4} .2735287{col 67}{space 3} .4712061
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) recastci(rarea) ///
> xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ///
> legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Aggregate Voluntary Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3B. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Aggregate Voluntary Turnover Model [MODEL 9]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram voluntary if voluntary<=0.15, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:voluntary}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3B.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3B.04-29-2025.gph} saved

{com}. *
. *
. ******************************************************************************************************************************************************************
. *********************************************************************************
. 
. * [MODEL 10, FIGURE 4B] H2: Turnover X Service Length
. eststo: xtgee delay_pct c.voluntary##c.volabove15_los15pct cio5 activityno_permil_pct lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 voluntaryb volabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb  , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.62314013
{txt}Iteration 2:  Tolerance = {res}.40319506
{txt}Iteration 3:  Tolerance = {res}.1583834
{txt}Iteration 4:  Tolerance = {res}.09699744
{txt}Iteration 5:  Tolerance = {res}.04821812
{txt}Iteration 6:  Tolerance = {res}.02365776
{txt}Iteration 7:  Tolerance = {res}.01108044
{txt}Iteration 8:  Tolerance = {res}.00510303
{txt}Iteration 9:  Tolerance = {res}.00232401
{txt}Iteration 10: Tolerance = {res}.00105256
{txt}Iteration 11: Tolerance = {res}.00047518
{txt}Iteration 12: Tolerance = {res}.00021412
{txt}Iteration 13: Tolerance = {res}.00009636
{txt}Iteration 14: Tolerance = {res}.00004333
{txt}Iteration 15: Tolerance = {res}.00001947
{txt}Iteration 16: Tolerance = {res}8.741e-06
{txt}Iteration 17: Tolerance = {res}3.924e-06
{txt}Iteration 18: Tolerance = {res}1.761e-06
{txt}Iteration 19: Tolerance = {res}7.897e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:375.80}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 99:(Std. err. adjusted for clustering on {res:a_id})}
{hline 34}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 35}{c |}{col 47}  Semirobust
{col 1}                        delay_pct{col 35}{c |} Coefficient{col 47}  std. err.{col 59}      z{col 67}   P>|z|{col 75}     [95% con{col 88}f. interval]
{hline 34}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 24}voluntary {c |}{col 35}{res}{space 2} 2.268077{col 47}{space 2} 1.261974{col 58}{space 1}    1.80{col 67}{space 3}0.072{col 75}{space 4} -.205347{col 88}{space 3} 4.741502
{txt}{space 14}volabove15_los15pct {c |}{col 35}{res}{space 2}-.1707308{col 47}{space 2} 2.094339{col 58}{space 1}   -0.08{col 67}{space 3}0.935{col 75}{space 4}-4.275559{col 88}{space 3} 3.934097
{txt}{space 33} {c |}
c.voluntary#c.volabove15_los15pct {c |}{col 35}{res}{space 2}-.3942163{col 47}{space 2} 12.45102{col 58}{space 1}   -0.03{col 67}{space 3}0.975{col 75}{space 4}-24.79777{col 88}{space 3} 24.00933
{txt}{space 33} {c |}
{space 29}cio5 {c |}{col 35}{res}{space 2}-.0254936{col 47}{space 2} .0935023{col 58}{space 1}   -0.27{col 67}{space 3}0.785{col 75}{space 4}-.2087547{col 88}{space 3} .1577675
{txt}{space 12}activityno_permil_pct {c |}{col 35}{res}{space 2} .2243588{col 47}{space 2} .2318939{col 58}{space 1}    0.97{col 67}{space 3}0.333{col 75}{space 4}-.2301449{col 88}{space 3} .6788625
{txt}{space 26}lnitemp {c |}{col 35}{res}{space 2}-.3885514{col 47}{space 2}  .278052{col 58}{space 1}   -1.40{col 67}{space 3}0.162{col 75}{space 4}-.9335233{col 88}{space 3} .1564205
{txt}{space 23}lnprojsize {c |}{col 35}{res}{space 2} .0390843{col 47}{space 2} .0292848{col 58}{space 1}    1.33{col 67}{space 3}0.182{col 75}{space 4}-.0183129{col 88}{space 3} .0964816
{txt}{space 22}projects_no {c |}{col 35}{res}{space 2} .0012328{col 47}{space 2} .0016494{col 58}{space 1}    0.75{col 67}{space 3}0.455{col 75}{space 4}    -.002{col 88}{space 3} .0044656
{txt}{space 26}accrate {c |}{col 35}{res}{space 2}-.0207741{col 47}{space 2} .5513999{col 58}{space 1}   -0.04{col 67}{space 3}0.970{col 75}{space 4}-1.101498{col 88}{space 3}  1.05995
{txt}{space 18}newprojects_pct {c |}{col 35}{res}{space 2}-.8617669{col 47}{space 2} .1661642{col 58}{space 1}   -5.19{col 67}{space 3}0.000{col 75}{space 4}-1.187443{col 88}{space 3} -.536091
{txt}{space 23}cio_tenure {c |}{col 35}{res}{space 2}-.0132838{col 47}{space 2} .0272955{col 58}{space 1}   -0.49{col 67}{space 3}0.626{col 75}{space 4}-.0667821{col 88}{space 3} .0402144
{txt}{space 31}y1 {c |}{col 35}{res}{space 2}-1.288284{col 47}{space 2} .0994384{col 58}{space 1}  -12.96{col 67}{space 3}0.000{col 75}{space 4} -1.48318{col 88}{space 3}-1.093388
{txt}{space 31}y2 {c |}{col 35}{res}{space 2}-1.219893{col 47}{space 2} .1220065{col 58}{space 1}  -10.00{col 67}{space 3}0.000{col 75}{space 4}-1.459021{col 88}{space 3}-.9807646
{txt}{space 31}y3 {c |}{col 35}{res}{space 2}-1.186451{col 47}{space 2} .1210098{col 58}{space 1}   -9.80{col 67}{space 3}0.000{col 75}{space 4}-1.423626{col 88}{space 3}-.9492767
{txt}{space 31}y4 {c |}{col 35}{res}{space 2}-.9768659{col 47}{space 2} .1274476{col 58}{space 1}   -7.66{col 67}{space 3}0.000{col 75}{space 4}-1.226659{col 88}{space 3}-.7270732
{txt}{space 31}y5 {c |}{col 35}{res}{space 2}-1.048939{col 47}{space 2} .1106628{col 58}{space 1}   -9.48{col 67}{space 3}0.000{col 75}{space 4}-1.265834{col 88}{space 3}-.8320441
{txt}{space 31}y6 {c |}{col 35}{res}{space 2}-1.023634{col 47}{space 2} .1136131{col 58}{space 1}   -9.01{col 67}{space 3}0.000{col 75}{space 4}-1.246312{col 88}{space 3}-.8009569
{txt}{space 31}y7 {c |}{col 35}{res}{space 2}        0{col 47}{txt}  (omitted)
{space 23}voluntaryb {c |}{col 35}{res}{space 2}-.2551732{col 47}{space 2} 2.781097{col 58}{space 1}   -0.09{col 67}{space 3}0.927{col 75}{space 4}-5.706024{col 88}{space 3} 5.195677
{txt}{space 13}volabove15_los15pctb {c |}{col 35}{res}{space 2}-7.575335{col 47}{space 2} 2.322111{col 58}{space 1}   -3.26{col 67}{space 3}0.001{col 75}{space 4}-12.12659{col 88}{space 3}-3.024082
{txt}{space 11}activityno_permil_pctb {c |}{col 35}{res}{space 2} .9141359{col 47}{space 2} .4139586{col 58}{space 1}    2.21{col 67}{space 3}0.027{col 75}{space 4}  .102792{col 88}{space 3}  1.72548
{txt}{space 25}lnitempb {c |}{col 35}{res}{space 2} .3814583{col 47}{space 2} .2835563{col 58}{space 1}    1.35{col 67}{space 3}0.179{col 75}{space 4}-.1743019{col 88}{space 3} .9372185
{txt}{space 22}lnprojsizeb {c |}{col 35}{res}{space 2}  .035422{col 47}{space 2} .0618135{col 58}{space 1}    0.57{col 67}{space 3}0.567{col 75}{space 4}-.0857302{col 88}{space 3} .1565742
{txt}{space 21}projects_nob {c |}{col 35}{res}{space 2}-.0053387{col 47}{space 2} .0029341{col 58}{space 1}   -1.82{col 67}{space 3}0.069{col 75}{space 4}-.0110893{col 88}{space 3}  .000412
{txt}{space 25}accrateb {c |}{col 35}{res}{space 2}-1.050944{col 47}{space 2} 1.290327{col 58}{space 1}   -0.81{col 67}{space 3}0.415{col 75}{space 4}-3.579938{col 88}{space 3}  1.47805
{txt}{space 17}newprojects_pctb {c |}{col 35}{res}{space 2}-1.102841{col 47}{space 2} .3337504{col 58}{space 1}   -3.30{col 67}{space 3}0.001{col 75}{space 4}-1.756979{col 88}{space 3}-.4487019
{txt}{space 22}cio_tenureb {c |}{col 35}{res}{space 2}-.0383883{col 47}{space 2} .0612835{col 58}{space 1}   -0.63{col 67}{space 3}0.531{col 75}{space 4}-.1585017{col 88}{space 3} .0817251
{txt}{space 11}dmeratio_totalspending {c |}{col 35}{res}{space 2} .0526139{col 47}{space 2} .3820945{col 58}{space 1}    0.14{col 67}{space 3}0.890{col 75}{space 4}-.6962776{col 88}{space 3} .8015053
{txt}{space 10}dmeratio_totalspendingb {c |}{col 35}{res}{space 2} .3991061{col 47}{space 2} .4678294{col 58}{space 1}    0.85{col 67}{space 3}0.394{col 75}{space 4}-.5178227{col 88}{space 3} 1.316035
{txt}{space 28}_cons {c |}{col 35}{res}{space 2} 1.112848{col 47}{space 2} .3092502{col 58}{space 1}    3.60{col 67}{space 3}0.000{col 75}{space 4} .5067285{col 88}{space 3} 1.718967
{txt}{hline 34}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est2{txt} stored)

{com}. 
. * Aggregate Voluntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins , dydx(voluntary) at(volabove15_los15pct=($vollos_p10 $vollos_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 3:.05}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2}-.0089816{col 26}{space 2} .1630486{col 37}{space 1}   -0.06{col 46}{space 3}0.956{col 54}{space 4} -.328551{col 67}{space 3} .3105878
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. margins , dydx(voluntary) at(volabove15_los15pct=(0(0.025)0.2))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:0}}
{lalign 7:2._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.025}}
{lalign 7:3._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.05}}
{lalign 7:4._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.075}}
{lalign 7:5._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.1}}
{lalign 7:6._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.125}}
{lalign 7:7._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.15}}
{lalign 7:8._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.175}}
{lalign 7:9._at: }{space 0}{lalign 16:volabove15_los~t} = {res:{ralign 4:.2}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .6777997{col 26}{space 2} .3835653{col 37}{space 1}    1.77{col 46}{space 3}0.077{col 54}{space 4}-.0739744{col 67}{space 3} 1.429574
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .6733058{col 26}{space 2} .3878815{col 37}{space 1}    1.74{col 46}{space 3}0.083{col 54}{space 4}-.0869279{col 67}{space 3}  1.43354
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .6688181{col 26}{space 2} .4086498{col 37}{space 1}    1.64{col 46}{space 3}0.102{col 54}{space 4}-.1321208{col 67}{space 3} 1.469757
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .6643367{col 26}{space 2} .4433303{col 37}{space 1}    1.50{col 46}{space 3}0.134{col 54}{space 4}-.2045747{col 67}{space 3} 1.533248
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .6598617{col 26}{space 2} .4887593{col 37}{space 1}    1.35{col 46}{space 3}0.177{col 54}{space 4}-.2980889{col 67}{space 3} 1.617812
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .6553934{col 26}{space 2} .5420513{col 37}{space 1}    1.21{col 46}{space 3}0.227{col 54}{space 4}-.4070077{col 67}{space 3} 1.717794
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .6509319{col 26}{space 2} .6009476{col 37}{space 1}    1.08{col 46}{space 3}0.279{col 54}{space 4}-.5269038{col 67}{space 3} 1.828767
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .6464773{col 26}{space 2} .6638041{col 37}{space 1}    0.97{col 46}{space 3}0.330{col 54}{space 4}-.6545548{col 67}{space 3} 1.947509
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .6420299{col 26}{space 2} .7294583{col 37}{space 1}    0.88{col 46}{space 3}0.379{col 54}{space 4}-.7876821{col 67}{space 3} 2.071742
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15" 0.2 "20" ,format(%9.0f)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departures' Service Length (>15 years, %)", margin(t=1)  size(11pt)) ///
> title("{c -(}bf: Figure 4B. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Aggregate Voluntary Turnover Model [MODEL 10]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram volabove15_los15pct if volabove15_los15pct<=0.2 , yaxis(2) xaxis(1)  fcolor(gray%20) yscale(axis(2) alt) lc(gray%20) xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15" 0.2 "20" ,format(%9.0f)) ylabel( ,format(%9.1f) angle(0) ) ylabel(,format(%9.0f) angle(0) axis(2)) ) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:volabove15_los15pct}{p_end}
{res}{txt}
{com}. 
. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4B.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4B.04-29-2025.gph} saved

{com}. 
. 
. ******************************************************************************************************************************************************************
. *********************************************************************************
. 
. * [MODEL 11, FIGURE 5B] H3: Turnover X Agency Centralization
. eststo: xtgee delay_pct c.voluntary##i.cio5 c.volabove15_los15pct activityno_permil_pct lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 voluntaryb volabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.68667215
{txt}Iteration 2:  Tolerance = {res}.10744855
{txt}Iteration 3:  Tolerance = {res}.03338691
{txt}Iteration 4:  Tolerance = {res}.00606521
{txt}Iteration 5:  Tolerance = {res}.00406501
{txt}Iteration 6:  Tolerance = {res}.00223796
{txt}Iteration 7:  Tolerance = {res}.0010927
{txt}Iteration 8:  Tolerance = {res}.00050083
{txt}Iteration 9:  Tolerance = {res}.00022103
{txt}Iteration 10: Tolerance = {res}.00009506
{txt}Iteration 11: Tolerance = {res}.00004013
{txt}Iteration 12: Tolerance = {res}.00001671
{txt}Iteration 13: Tolerance = {res}6.880e-06
{txt}Iteration 14: Tolerance = {res}2.810e-06
{txt}Iteration 15: Tolerance = {res}1.140e-06
{txt}Iteration 16: Tolerance = {res}4.597e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:381.27}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 14}voluntary {c |}{col 25}{res}{space 2} 2.868439{col 37}{space 2} 1.330624{col 48}{space 1}    2.16{col 57}{space 3}0.031{col 65}{space 4} .2604652{col 78}{space 3} 5.476414
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}  .127943{col 37}{space 2} .1291196{col 48}{space 1}    0.99{col 57}{space 3}0.322{col 65}{space 4}-.1251268{col 78}{space 3} .3810129
{txt}{space 23} {c |}
{space 7}cio5#c.voluntary {c |}
{space 21}1  {c |}{col 25}{res}{space 2} -3.99106{col 37}{space 2} 1.933847{col 48}{space 1}   -2.06{col 57}{space 3}0.039{col 65}{space 4} -7.78133{col 78}{space 3}-.2007895
{txt}{space 23} {c |}
{space 4}volabove15_los15pct {c |}{col 25}{res}{space 2}-.2032437{col 37}{space 2} 1.037023{col 48}{space 1}   -0.20{col 57}{space 3}0.845{col 65}{space 4}-2.235771{col 78}{space 3} 1.829284
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2426815{col 37}{space 2} .2262925{col 48}{space 1}    1.07{col 57}{space 3}0.284{col 65}{space 4}-.2008437{col 78}{space 3} .6862066
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.3555531{col 37}{space 2} .2778018{col 48}{space 1}   -1.28{col 57}{space 3}0.201{col 65}{space 4}-.9000346{col 78}{space 3} .1889284
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2}  .046159{col 37}{space 2} .0280642{col 48}{space 1}    1.64{col 57}{space 3}0.100{col 65}{space 4}-.0088458{col 78}{space 3} .1011638
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0009645{col 37}{space 2} .0016032{col 48}{space 1}    0.60{col 57}{space 3}0.547{col 65}{space 4}-.0021778{col 78}{space 3} .0041067
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} -.136967{col 37}{space 2} .5390963{col 48}{space 1}   -0.25{col 57}{space 3}0.799{col 65}{space 4}-1.193576{col 78}{space 3} .9196423
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2} -.854798{col 37}{space 2} .1650824{col 48}{space 1}   -5.18{col 57}{space 3}0.000{col 65}{space 4}-1.178354{col 78}{space 3}-.5312423
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0169931{col 37}{space 2}  .026516{col 48}{space 1}   -0.64{col 57}{space 3}0.522{col 65}{space 4}-.0689636{col 78}{space 3} .0349774
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.293378{col 37}{space 2}  .099807{col 48}{space 1}  -12.96{col 57}{space 3}0.000{col 65}{space 4}-1.488996{col 78}{space 3}-1.097759
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.217106{col 37}{space 2} .1222861{col 48}{space 1}   -9.95{col 57}{space 3}0.000{col 65}{space 4}-1.456782{col 78}{space 3}-.9774297
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.183774{col 37}{space 2} .1197241{col 48}{space 1}   -9.89{col 57}{space 3}0.000{col 65}{space 4}-1.418429{col 78}{space 3}-.9491194
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.9959826{col 37}{space 2} .1265255{col 48}{space 1}   -7.87{col 57}{space 3}0.000{col 65}{space 4}-1.243968{col 78}{space 3}-.7479971
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-1.038574{col 37}{space 2}  .109815{col 48}{space 1}   -9.46{col 57}{space 3}0.000{col 65}{space 4}-1.253808{col 78}{space 3}-.8233407
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-1.013052{col 37}{space 2} .1100773{col 48}{space 1}   -9.20{col 57}{space 3}0.000{col 65}{space 4}  -1.2288{col 78}{space 3}-.7973047
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 13}voluntaryb {c |}{col 25}{res}{space 2}-.3723774{col 37}{space 2}  2.76783{col 48}{space 1}   -0.13{col 57}{space 3}0.893{col 65}{space 4}-5.797224{col 78}{space 3} 5.052469
{txt}{space 3}volabove15_los15pctb {c |}{col 25}{res}{space 2}-7.638682{col 37}{space 2} 2.349248{col 48}{space 1}   -3.25{col 57}{space 3}0.001{col 65}{space 4}-12.24312{col 78}{space 3}-3.034241
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9071423{col 37}{space 2} .4121568{col 48}{space 1}    2.20{col 57}{space 3}0.028{col 65}{space 4} .0993297{col 78}{space 3} 1.714955
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .3484917{col 37}{space 2} .2831023{col 48}{space 1}    1.23{col 57}{space 3}0.218{col 65}{space 4}-.2063787{col 78}{space 3} .9033621
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0305716{col 37}{space 2} .0604764{col 48}{space 1}    0.51{col 57}{space 3}0.613{col 65}{space 4}  -.08796{col 78}{space 3} .1491032
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2} -.005355{col 37}{space 2} .0028598{col 48}{space 1}   -1.87{col 57}{space 3}0.061{col 65}{space 4}-.0109601{col 78}{space 3} .0002502
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-.9307788{col 37}{space 2} 1.281936{col 48}{space 1}   -0.73{col 57}{space 3}0.468{col 65}{space 4}-3.443327{col 78}{space 3} 1.581769
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.085267{col 37}{space 2} .3366973{col 48}{space 1}   -3.22{col 57}{space 3}0.001{col 65}{space 4}-1.745181{col 78}{space 3} -.425352
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0314033{col 37}{space 2} .0624041{col 48}{space 1}   -0.50{col 57}{space 3}0.615{col 65}{space 4}-.1537131{col 78}{space 3} .0909065
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .0307867{col 37}{space 2} .3794899{col 48}{space 1}    0.08{col 57}{space 3}0.935{col 65}{space 4}-.7129998{col 78}{space 3} .7745732
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .4577344{col 37}{space 2} .4656674{col 48}{space 1}    0.98{col 57}{space 3}0.326{col 65}{space 4}-.4549569{col 78}{space 3} 1.370426
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} 1.061442{col 37}{space 2} .3239997{col 48}{space 1}    3.28{col 57}{space 3}0.001{col 65}{space 4} .4264144{col 78}{space 3}  1.69647
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est3{txt} stored)

{com}. 
. * Aggregate Voluntary Turnover Rate: Marginal Effect Differentials Based on Agency Centralization
. margins , dydx(voluntary) over(cio5) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 8}cio5 {c |}
{space 5}1 vs 0  {c |}{col 14}{res}{space 2}-1.186576{col 26}{space 2} .5689017{col 37}{space 1}   -2.09{col 46}{space 3}0.037{col 54}{space 4}-2.301603{col 67}{space 3}-.0715493
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. 
. margins , dydx(voluntary) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 8}cio5 {c |}
{space 10}0  {c |}{col 14}{res}{space 2} .8579497{col 26}{space 2} .3993069{col 37}{space 1}    2.15{col 46}{space 3}0.032{col 54}{space 4} .0753226{col 67}{space 3} 1.640577
{txt}{space 10}1  {c |}{col 14}{res}{space 2}-.3286263{col 26}{space 2} .5998778{col 37}{space 1}   -0.55{col 46}{space 3}0.584{col 54}{space 4}-1.504365{col 67}{space 3} .8471126
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(black)) ciopt(color(black%60))  yline(0, lcolor(red) lpattern(shortdash)) ///
> xlabel (-1 " " 0 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5B. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Aggregate Voluntary Turnover Model [MODEL 11]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5B.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5B.04-29-2025.gph} saved

{com}. 
. *
. *
. * ******************************************************************************************************************************************************************
. *********************************************************************************
. 
. * [MODEL 12, FIGURE 6B] H4 STANDARDIZATION HYPOTHESIS
. eststo: xtgee delay_pct c.voluntary##c.activityno_permil_pct c.volabove15_los15pct  cio5 lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 voluntaryb volabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.65101209
{txt}Iteration 2:  Tolerance = {res}.11007329
{txt}Iteration 3:  Tolerance = {res}.03232733
{txt}Iteration 4:  Tolerance = {res}.00755324
{txt}Iteration 5:  Tolerance = {res}.0037395
{txt}Iteration 6:  Tolerance = {res}.00172452
{txt}Iteration 7:  Tolerance = {res}.00087193
{txt}Iteration 8:  Tolerance = {res}.0004246
{txt}Iteration 9:  Tolerance = {res}.00020124
{txt}Iteration 10: Tolerance = {res}.00009364
{txt}Iteration 11: Tolerance = {res}.00004296
{txt}Iteration 12: Tolerance = {res}.0000195
{txt}Iteration 13: Tolerance = {res}8.780e-06
{txt}Iteration 14: Tolerance = {res}3.928e-06
{txt}Iteration 15: Tolerance = {res}1.749e-06
{txt}Iteration 16: Tolerance = {res}7.759e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:402.71}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 101:(Std. err. adjusted for clustering on {res:a_id})}
{hline 36}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 37}{c |}{col 49}  Semirobust
{col 1}                          delay_pct{col 37}{c |} Coefficient{col 49}  std. err.{col 61}      z{col 69}   P>|z|{col 77}     [95% con{col 90}f. interval]
{hline 36}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}voluntary {c |}{col 37}{res}{space 2} 3.172659{col 49}{space 2} 1.804065{col 60}{space 1}    1.76{col 69}{space 3}0.079{col 77}{space 4}-.3632427{col 90}{space 3} 6.708561
{txt}{space 14}activityno_permil_pct {c |}{col 37}{res}{space 2}  .317448{col 49}{space 2} .2683083{col 60}{space 1}    1.18{col 69}{space 3}0.237{col 77}{space 4}-.2084266{col 90}{space 3} .8433227
{txt}{space 35} {c |}
c.voluntary#c.activityno_permil_pct {c |}{col 37}{res}{space 2}-2.309759{col 49}{space 2} 2.523712{col 60}{space 1}   -0.92{col 69}{space 3}0.360{col 77}{space 4}-7.256144{col 90}{space 3} 2.636626
{txt}{space 35} {c |}
{space 16}volabove15_los15pct {c |}{col 37}{res}{space 2}-.3300247{col 49}{space 2} 1.047856{col 60}{space 1}   -0.31{col 69}{space 3}0.753{col 77}{space 4}-2.383785{col 90}{space 3} 1.723735
{txt}{space 31}cio5 {c |}{col 37}{res}{space 2} -.018217{col 49}{space 2} .0934019{col 60}{space 1}   -0.20{col 69}{space 3}0.845{col 77}{space 4}-.2012814{col 90}{space 3} .1648475
{txt}{space 28}lnitemp {c |}{col 37}{res}{space 2}-.3839151{col 49}{space 2} .2806963{col 60}{space 1}   -1.37{col 69}{space 3}0.171{col 77}{space 4}-.9340698{col 90}{space 3} .1662395
{txt}{space 25}lnprojsize {c |}{col 37}{res}{space 2}    .0399{col 49}{space 2} .0291816{col 60}{space 1}    1.37{col 69}{space 3}0.172{col 77}{space 4}-.0172949{col 90}{space 3} .0970948
{txt}{space 24}projects_no {c |}{col 37}{res}{space 2} .0013542{col 49}{space 2} .0016352{col 60}{space 1}    0.83{col 69}{space 3}0.408{col 77}{space 4}-.0018509{col 90}{space 3} .0045592
{txt}{space 28}accrate {c |}{col 37}{res}{space 2}-.0602146{col 49}{space 2} .5362366{col 60}{space 1}   -0.11{col 69}{space 3}0.911{col 77}{space 4}-1.111219{col 90}{space 3} .9907897
{txt}{space 20}newprojects_pct {c |}{col 37}{res}{space 2}-.8706675{col 49}{space 2} .1635787{col 60}{space 1}   -5.32{col 69}{space 3}0.000{col 77}{space 4}-1.191276{col 90}{space 3}-.5500592
{txt}{space 25}cio_tenure {c |}{col 37}{res}{space 2}-.0116097{col 49}{space 2} .0267823{col 60}{space 1}   -0.43{col 69}{space 3}0.665{col 77}{space 4} -.064102{col 90}{space 3} .0408825
{txt}{space 33}y1 {c |}{col 37}{res}{space 2}-1.297053{col 49}{space 2} .0989784{col 60}{space 1}  -13.10{col 69}{space 3}0.000{col 77}{space 4}-1.491047{col 90}{space 3}-1.103059
{txt}{space 33}y2 {c |}{col 37}{res}{space 2}-1.228252{col 49}{space 2} .1219116{col 60}{space 1}  -10.07{col 69}{space 3}0.000{col 77}{space 4}-1.467194{col 90}{space 3}-.9893092
{txt}{space 33}y3 {c |}{col 37}{res}{space 2}-1.196759{col 49}{space 2} .1207251{col 60}{space 1}   -9.91{col 69}{space 3}0.000{col 77}{space 4}-1.433376{col 90}{space 3}-.9601422
{txt}{space 33}y4 {c |}{col 37}{res}{space 2}-.9846008{col 49}{space 2} .1269653{col 60}{space 1}   -7.75{col 69}{space 3}0.000{col 77}{space 4}-1.233448{col 90}{space 3}-.7357534
{txt}{space 33}y5 {c |}{col 37}{res}{space 2}-1.058209{col 49}{space 2} .1103263{col 60}{space 1}   -9.59{col 69}{space 3}0.000{col 77}{space 4}-1.274444{col 90}{space 3}-.8419731
{txt}{space 33}y6 {c |}{col 37}{res}{space 2}-1.025243{col 49}{space 2} .1121308{col 60}{space 1}   -9.14{col 69}{space 3}0.000{col 77}{space 4}-1.245015{col 90}{space 3}-.8054703
{txt}{space 33}y7 {c |}{col 37}{res}{space 2}        0{col 49}{txt}  (omitted)
{space 25}voluntaryb {c |}{col 37}{res}{space 2}-.3487195{col 49}{space 2}  2.80227{col 60}{space 1}   -0.12{col 69}{space 3}0.901{col 77}{space 4}-5.841069{col 90}{space 3}  5.14363
{txt}{space 15}volabove15_los15pctb {c |}{col 37}{res}{space 2}-7.623417{col 49}{space 2} 2.291015{col 60}{space 1}   -3.33{col 69}{space 3}0.001{col 77}{space 4}-12.11372{col 90}{space 3} -3.13311
{txt}{space 13}activityno_permil_pctb {c |}{col 37}{res}{space 2} .9223394{col 49}{space 2} .4087111{col 60}{space 1}    2.26{col 69}{space 3}0.024{col 77}{space 4} .1212803{col 90}{space 3} 1.723399
{txt}{space 27}lnitempb {c |}{col 37}{res}{space 2} .3769937{col 49}{space 2} .2869098{col 60}{space 1}    1.31{col 69}{space 3}0.189{col 77}{space 4}-.1853391{col 90}{space 3} .9393266
{txt}{space 24}lnprojsizeb {c |}{col 37}{res}{space 2} .0342893{col 49}{space 2} .0612111{col 60}{space 1}    0.56{col 69}{space 3}0.575{col 77}{space 4}-.0856822{col 90}{space 3} .1542607
{txt}{space 23}projects_nob {c |}{col 37}{res}{space 2}-.0055114{col 49}{space 2}  .002906{col 60}{space 1}   -1.90{col 69}{space 3}0.058{col 77}{space 4}-.0112071{col 90}{space 3} .0001843
{txt}{space 27}accrateb {c |}{col 37}{res}{space 2} -1.06245{col 49}{space 2} 1.249943{col 60}{space 1}   -0.85{col 69}{space 3}0.395{col 77}{space 4}-3.512293{col 90}{space 3} 1.387393
{txt}{space 19}newprojects_pctb {c |}{col 37}{res}{space 2}-1.086243{col 49}{space 2} .3343992{col 60}{space 1}   -3.25{col 69}{space 3}0.001{col 77}{space 4}-1.741654{col 90}{space 3}-.4308331
{txt}{space 24}cio_tenureb {c |}{col 37}{res}{space 2}-.0395195{col 49}{space 2}  .061585{col 60}{space 1}   -0.64{col 69}{space 3}0.521{col 77}{space 4}-.1602238{col 90}{space 3} .0811849
{txt}{space 13}dmeratio_totalspending {c |}{col 37}{res}{space 2} .0822923{col 49}{space 2} .3855882{col 60}{space 1}    0.21{col 69}{space 3}0.831{col 77}{space 4}-.6734467{col 90}{space 3} .8380313
{txt}{space 12}dmeratio_totalspendingb {c |}{col 37}{res}{space 2} .3943639{col 49}{space 2} .4714881{col 60}{space 1}    0.84{col 69}{space 3}0.403{col 77}{space 4}-.5297357{col 90}{space 3} 1.318464
{txt}{space 30}_cons {c |}{col 37}{res}{space 2} 1.082041{col 49}{space 2} .3140035{col 60}{space 1}    3.45{col 69}{space 3}0.001{col 77}{space 4}  .466606{col 90}{space 3} 1.697477
{txt}{hline 36}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est4{txt} stored)

{com}. 
. * Aggregate Voluntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Task Standardization
. margins , dydx(voluntary) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2}-.4381081{col 26}{space 2} .5225256{col 37}{space 1}   -0.84{col 46}{space 3}0.402{col 54}{space 4}-1.462239{col 67}{space 3} .5860233
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 6B] Conditional Turnover Effects by Task Standardization (H4)
. margins , dydx(voluntary) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary    {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .9155886{col 26}{space 2} .5267954{col 37}{space 1}    1.74{col 46}{space 3}0.082{col 54}{space 4}-.1169114{col 67}{space 3} 1.948089
{txt}{space 10}2  {c |}{col 14}{res}{space 2}  .859157{col 26}{space 2}  .483332{col 37}{space 1}    1.78{col 46}{space 3}0.075{col 54}{space 4}-.0881563{col 67}{space 3}  1.80647
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .8009228{col 26}{space 2} .4451764{col 37}{space 1}    1.80{col 46}{space 3}0.072{col 54}{space 4} -.071607{col 67}{space 3} 1.673453
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .7409247{col 26}{space 2} .4148224{col 37}{space 1}    1.79{col 46}{space 3}0.074{col 54}{space 4}-.0721123{col 67}{space 3} 1.553962
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .6792059{col 26}{space 2} .3951707{col 37}{space 1}    1.72{col 46}{space 3}0.086{col 54}{space 4}-.0953144{col 67}{space 3} 1.453726
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .6158139{col 26}{space 2} .3889682{col 37}{space 1}    1.58{col 46}{space 3}0.113{col 54}{space 4}-.1465497{col 67}{space 3} 1.378177
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .5508006{col 26}{space 2} .3979267{col 37}{space 1}    1.38{col 46}{space 3}0.166{col 54}{space 4}-.2291214{col 67}{space 3} 1.330723
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .4842222{col 26}{space 2} .4220728{col 37}{space 1}    1.15{col 46}{space 3}0.251{col 54}{space 4}-.3430254{col 67}{space 3}  1.31147
{txt}{space 10}9  {c |}{col 14}{res}{space 2}  .416139{col 26}{space 2} .4598886{col 37}{space 1}    0.90{col 46}{space 3}0.366{col 54}{space 4} -.485226{col 67}{space 3} 1.317504
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .3466156{col 26}{space 2} .5090716{col 37}{space 1}    0.68{col 46}{space 3}0.496{col 54}{space 4}-.6511464{col 67}{space 3} 1.344377
{txt}{space 9}11  {c |}{col 14}{res}{space 2} .2757205{col 26}{space 2} .5672769{col 37}{space 1}    0.49{col 46}{space 3}0.627{col 54}{space 4}-.8361219{col 67}{space 3} 1.387563
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", size(11pt) margin(t=1)) ///
> title("{c -(}bf: Figure 6B. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Aggregate Voluntary Turnover Model [MODEL 12]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0))  ylabel( ,format(%9.1f) angle(0) axis(2)) ytitle("",  axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6B.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6B.04-29-2025.gph} saved

{com}. 
. 
. 
. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. **# [MODELS 13 - 16, FIGURES 3C - 6C] AGGREGATE INVOLUNTARY TURNOVER MODELS *
. 
. * [MODEL 13, FIGURE 3C] H1: Aggregate Involuntary Turnover Baseline Effects
. eststo: xtgee delay_pct c.involuntary c.activityno_permil_pct c.involabove15_los15pct i.cio5 lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 involuntaryb involabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}3.517617
{txt}Iteration 2:  Tolerance = {res}.22620583
{txt}Iteration 3:  Tolerance = {res}.0656624
{txt}Iteration 4:  Tolerance = {res}.05394148
{txt}Iteration 5:  Tolerance = {res}.02325447
{txt}Iteration 6:  Tolerance = {res}.00923885
{txt}Iteration 7:  Tolerance = {res}.00349142
{txt}Iteration 8:  Tolerance = {res}.0012241
{txt}Iteration 9:  Tolerance = {res}.00055318
{txt}Iteration 10: Tolerance = {res}.00032341
{txt}Iteration 11: Tolerance = {res}.00018058
{txt}Iteration 12: Tolerance = {res}.00009772
{txt}Iteration 13: Tolerance = {res}.00005169
{txt}Iteration 14: Tolerance = {res}.00002688
{txt}Iteration 15: Tolerance = {res}.00001379
{txt}Iteration 16: Tolerance = {res}7.003e-06
{txt}Iteration 17: Tolerance = {res}3.524e-06
{txt}Iteration 18: Tolerance = {res}1.761e-06
{txt}Iteration 19: Tolerance = {res}8.741e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:330.76}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 12}involuntary {c |}{col 25}{res}{space 2}-3.327305{col 37}{space 2} 2.946507{col 48}{space 1}   -1.13{col 57}{space 3}0.259{col 65}{space 4}-9.102352{col 78}{space 3} 2.447742
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2396971{col 37}{space 2} .2221233{col 48}{space 1}    1.08{col 57}{space 3}0.281{col 65}{space 4}-.1956564{col 78}{space 3} .6750507
{txt}{space 2}involabove15_los15pct {c |}{col 25}{res}{space 2} 2.458767{col 37}{space 2} 11.81075{col 48}{space 1}    0.21{col 57}{space 3}0.835{col 65}{space 4}-20.68988{col 78}{space 3} 25.60742
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.0619108{col 37}{space 2} .0972908{col 48}{space 1}   -0.64{col 57}{space 3}0.525{col 65}{space 4}-.2525973{col 78}{space 3} .1287757
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.4048955{col 37}{space 2} .3076178{col 48}{space 1}   -1.32{col 57}{space 3}0.188{col 65}{space 4}-1.007815{col 78}{space 3} .1980243
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0310855{col 37}{space 2} .0300028{col 48}{space 1}    1.04{col 57}{space 3}0.300{col 65}{space 4}-.0277189{col 78}{space 3} .0898898
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0012736{col 37}{space 2} .0016426{col 48}{space 1}    0.78{col 57}{space 3}0.438{col 65}{space 4}-.0019459{col 78}{space 3}  .004493
{txt}{space 16}accrate {c |}{col 25}{res}{space 2}  .381559{col 37}{space 2} .6283441{col 48}{space 1}    0.61{col 57}{space 3}0.544{col 65}{space 4}-.8499728{col 78}{space 3} 1.613091
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2} -.790989{col 37}{space 2}  .175193{col 48}{space 1}   -4.51{col 57}{space 3}0.000{col 65}{space 4}-1.134361{col 78}{space 3}-.4476169
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0150021{col 37}{space 2} .0274858{col 48}{space 1}   -0.55{col 57}{space 3}0.585{col 65}{space 4}-.0688732{col 78}{space 3}  .038869
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.231617{col 37}{space 2} .1022053{col 48}{space 1}  -12.05{col 57}{space 3}0.000{col 65}{space 4}-1.431936{col 78}{space 3}-1.031298
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.185982{col 37}{space 2} .1224073{col 48}{space 1}   -9.69{col 57}{space 3}0.000{col 65}{space 4}-1.425896{col 78}{space 3}-.9460684
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.140292{col 37}{space 2} .1255585{col 48}{space 1}   -9.08{col 57}{space 3}0.000{col 65}{space 4}-1.386382{col 78}{space 3}-.8942017
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.9078776{col 37}{space 2} .1297656{col 48}{space 1}   -7.00{col 57}{space 3}0.000{col 65}{space 4}-1.162213{col 78}{space 3}-.6535417
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9923367{col 37}{space 2}  .116424{col 48}{space 1}   -8.52{col 57}{space 3}0.000{col 65}{space 4}-1.220524{col 78}{space 3}-.7641498
{txt}{space 21}y6 {c |}{col 25}{res}{space 2} -1.00057{col 37}{space 2} .1126299{col 48}{space 1}   -8.88{col 57}{space 3}0.000{col 65}{space 4}-1.221321{col 78}{space 3}-.7798197
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 11}involuntaryb {c |}{col 25}{res}{space 2}   7.3837{col 37}{space 2} 5.435803{col 48}{space 1}    1.36{col 57}{space 3}0.174{col 65}{space 4}-3.270279{col 78}{space 3} 18.03768
{txt}{space 1}involabove15_los15pctb {c |}{col 25}{res}{space 2}-19.23996{col 37}{space 2} 29.05161{col 48}{space 1}   -0.66{col 57}{space 3}0.508{col 65}{space 4}-76.18006{col 78}{space 3} 37.70015
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8047978{col 37}{space 2} .3935487{col 48}{space 1}    2.04{col 57}{space 3}0.041{col 65}{space 4} .0334565{col 78}{space 3} 1.576139
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .4446658{col 37}{space 2} .3175665{col 48}{space 1}    1.40{col 57}{space 3}0.161{col 65}{space 4}-.1777531{col 78}{space 3} 1.067085
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0066063{col 37}{space 2} .0696539{col 48}{space 1}    0.09{col 57}{space 3}0.924{col 65}{space 4}-.1299129{col 78}{space 3} .1431255
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0038223{col 37}{space 2} .0026599{col 48}{space 1}   -1.44{col 57}{space 3}0.151{col 65}{space 4}-.0090355{col 78}{space 3} .0013909
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-2.044924{col 37}{space 2} 1.452002{col 48}{space 1}   -1.41{col 57}{space 3}0.159{col 65}{space 4}-4.890794{col 78}{space 3} .8009469
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.066728{col 37}{space 2} .3755204{col 48}{space 1}   -2.84{col 57}{space 3}0.005{col 65}{space 4}-1.802734{col 78}{space 3}-.3307215
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0185494{col 37}{space 2} .0628499{col 48}{space 1}   -0.30{col 57}{space 3}0.768{col 65}{space 4}-.1417329{col 78}{space 3} .1046341
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2}-.0424662{col 37}{space 2} .3948138{col 48}{space 1}   -0.11{col 57}{space 3}0.914{col 65}{space 4}-.8162871{col 78}{space 3} .7313546
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3630645{col 37}{space 2} .5043325{col 48}{space 1}    0.72{col 57}{space 3}0.472{col 65}{space 4}-.6254089{col 78}{space 3} 1.351538
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .8208514{col 37}{space 2} .4214627{col 48}{space 1}    1.95{col 57}{space 3}0.051{col 65}{space 4}-.0052003{col 78}{space 3} 1.646903
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est5{txt} stored)

{com}. 
. * Aggregate Involuntary Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(involuntary)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 1}involuntary {c |}{col 14}{res}{space 2}-1.006894{col 26}{space 2} .8911044{col 37}{space 1}   -1.13{col 46}{space 3}0.259{col 54}{space 4}-2.753427{col 67}{space 3} .7396387
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Aggregate Involuntary Turnover Rate
. margins, at(involuntary=($involuntary_p10 $involuntary_p90)) pwcompare(effects)
{res}
{txt}{col 1}Pairwise comparisons of predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 3:.01}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}   Contrast{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} -.010041{col 26}{space 2} .0088592{col 37}{space 1}   -1.13{col 46}{space 3}0.257{col 54}{space 4}-.0274048{col 67}{space 3} .0073228
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3C]
. margins, at(involuntary=(0(0.005)0.07)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:0}}
{lalign 8:2._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.005}}
{lalign 8:3._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.01}}
{lalign 8:4._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.015}}
{lalign 8:5._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.02}}
{lalign 8:6._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.025}}
{lalign 8:7._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.03}}
{lalign 8:8._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.035}}
{lalign 8:9._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.04}}
{lalign 8:10._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.045}}
{lalign 8:11._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.05}}
{lalign 8:12._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.055}}
{lalign 8:13._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.06}}
{lalign 8:14._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.065}}
{lalign 8:15._at: }{space 0}{lalign 11:involuntary} = {res:{ralign 4:.07}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2983622{col 26}{space 2} .0151471{col 37}{space 1}   19.70{col 46}{space 3}0.000{col 54}{space 4} .2686744{col 67}{space 3} .3280499
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .2933206{col 26}{space 2} .0147262{col 37}{space 1}   19.92{col 46}{space 3}0.000{col 54}{space 4} .2644578{col 67}{space 3} .3221834
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .2883212{col 26}{space 2} .0155977{col 37}{space 1}   18.48{col 46}{space 3}0.000{col 54}{space 4} .2577503{col 67}{space 3} .3188921
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2833646{col 26}{space 2} .0175133{col 37}{space 1}   16.18{col 46}{space 3}0.000{col 54}{space 4} .2490393{col 67}{space 3}   .31769
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2784516{col 26}{space 2} .0201275{col 37}{space 1}   13.83{col 46}{space 3}0.000{col 54}{space 4} .2390025{col 67}{space 3} .3179007
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .2735828{col 26}{space 2} .0231613{col 37}{space 1}   11.81{col 46}{space 3}0.000{col 54}{space 4} .2281876{col 67}{space 3} .3189781
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .2687588{col 26}{space 2} .0264319{col 37}{space 1}   10.17{col 46}{space 3}0.000{col 54}{space 4} .2169531{col 67}{space 3} .3205644
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .2639801{col 26}{space 2} .0298271{col 37}{space 1}    8.85{col 46}{space 3}0.000{col 54}{space 4}   .20552{col 67}{space 3} .3224402
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .2592474{col 26}{space 2} .0332779{col 37}{space 1}    7.79{col 46}{space 3}0.000{col 54}{space 4} .1940239{col 67}{space 3} .3244708
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .2545611{col 26}{space 2} .0367405{col 37}{space 1}    6.93{col 46}{space 3}0.000{col 54}{space 4}  .182551{col 67}{space 3} .3265712
{txt}{space 9}11  {c |}{col 14}{res}{space 2} .2499218{col 26}{space 2} .0401865{col 37}{space 1}    6.22{col 46}{space 3}0.000{col 54}{space 4} .1711576{col 67}{space 3} .3286859
{txt}{space 9}12  {c |}{col 14}{res}{space 2} .2453299{col 26}{space 2} .0435967{col 37}{space 1}    5.63{col 46}{space 3}0.000{col 54}{space 4}  .159882{col 67}{space 3} .3307778
{txt}{space 9}13  {c |}{col 14}{res}{space 2}  .240786{col 26}{space 2} .0469573{col 37}{space 1}    5.13{col 46}{space 3}0.000{col 54}{space 4} .1487513{col 67}{space 3} .3328206
{txt}{space 9}14  {c |}{col 14}{res}{space 2} .2362904{col 26}{space 2} .0502588{col 37}{space 1}    4.70{col 46}{space 3}0.000{col 54}{space 4} .1377849{col 67}{space 3} .3347958
{txt}{space 9}15  {c |}{col 14}{res}{space 2} .2318435{col 26}{space 2} .0534939{col 37}{space 1}    4.33{col 46}{space 3}0.000{col 54}{space 4} .1269974{col 67}{space 3} .3366895
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) recastci(rarea) ///
> xlabel(0 "0" 0.02 "2" 0.04 "4" 0.06 "6", format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ///
> legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Aggregate Involuntary Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3C. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Aggregate Involuntary Turnover Model [MODEL 13]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram involuntary if involuntary<=0.07, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.02 "2" 0.04 "4" 0.06 "6",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:involuntary}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3C.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3C.04-29-2025.gph} saved

{com}. 
. ***************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 14, FIGURE 4C] H2: Turnover X Service Length 
. eststo: xtgee delay_pct c.involuntary##c.involabove15_los15pct c.activityno_permil_pct  i.cio5 lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 involuntaryb involabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}3.3479573
{txt}Iteration 2:  Tolerance = {res}.11842425
{txt}Iteration 3:  Tolerance = {res}.05417242
{txt}Iteration 4:  Tolerance = {res}.01530409
{txt}Iteration 5:  Tolerance = {res}.00674362
{txt}Iteration 6:  Tolerance = {res}.00427312
{txt}Iteration 7:  Tolerance = {res}.00417216
{txt}Iteration 8:  Tolerance = {res}.00307583
{txt}Iteration 9:  Tolerance = {res}.00200112
{txt}Iteration 10: Tolerance = {res}.00121242
{txt}Iteration 11: Tolerance = {res}.00070151
{txt}Iteration 12: Tolerance = {res}.00039289
{txt}Iteration 13: Tolerance = {res}.0002147
{txt}Iteration 14: Tolerance = {res}.00011506
{txt}Iteration 15: Tolerance = {res}.00006068
{txt}Iteration 16: Tolerance = {res}.00003156
{txt}Iteration 17: Tolerance = {res}.00001623
{txt}Iteration 18: Tolerance = {res}8.252e-06
{txt}Iteration 19: Tolerance = {res}4.156e-06
{txt}Iteration 20: Tolerance = {res}2.075e-06
{txt}Iteration 21: Tolerance = {res}1.027e-06
{txt}Iteration 22: Tolerance = {res}5.039e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:467.62}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 103:(Std. err. adjusted for clustering on {res:a_id})}
{hline 38}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 39}{c |}{col 51}  Semirobust
{col 1}                            delay_pct{col 39}{c |} Coefficient{col 51}  std. err.{col 63}      z{col 71}   P>|z|{col 79}     [95% con{col 92}f. interval]
{hline 38}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}involuntary {c |}{col 39}{res}{space 2} -5.45469{col 51}{space 2} 3.253213{col 62}{space 1}   -1.68{col 71}{space 3}0.094{col 79}{space 4}-11.83087{col 92}{space 3} .9214905
{txt}{space 16}involabove15_los15pct {c |}{col 39}{res}{space 2}-9.026737{col 51}{space 2} 16.24114{col 62}{space 1}   -0.56{col 71}{space 3}0.578{col 79}{space 4}-40.85878{col 92}{space 3} 22.80531
{txt}{space 37} {c |}
c.involuntary#c.involabove15_los15pct {c |}{col 39}{res}{space 2} 185.6027{col 51}{space 2} 114.7277{col 62}{space 1}    1.62{col 71}{space 3}0.106{col 79}{space 4}-39.25953{col 92}{space 3} 410.4649
{txt}{space 37} {c |}
{space 16}activityno_permil_pct {c |}{col 39}{res}{space 2} .2340496{col 51}{space 2} .2254839{col 62}{space 1}    1.04{col 71}{space 3}0.299{col 79}{space 4}-.2078906{col 92}{space 3} .6759898
{txt}{space 31}1.cio5 {c |}{col 39}{res}{space 2}-.0530613{col 51}{space 2} .0996389{col 62}{space 1}   -0.53{col 71}{space 3}0.594{col 79}{space 4}-.2483499{col 92}{space 3} .1422273
{txt}{space 30}lnitemp {c |}{col 39}{res}{space 2}-.4404005{col 51}{space 2} .3031752{col 62}{space 1}   -1.45{col 71}{space 3}0.146{col 79}{space 4}-1.034613{col 92}{space 3} .1538119
{txt}{space 27}lnprojsize {c |}{col 39}{res}{space 2} .0286361{col 51}{space 2} .0303965{col 62}{space 1}    0.94{col 71}{space 3}0.346{col 79}{space 4}  -.03094{col 92}{space 3} .0882123
{txt}{space 26}projects_no {c |}{col 39}{res}{space 2} .0012458{col 51}{space 2} .0016999{col 62}{space 1}    0.73{col 71}{space 3}0.464{col 79}{space 4} -.002086{col 92}{space 3} .0045775
{txt}{space 30}accrate {c |}{col 39}{res}{space 2} .3405255{col 51}{space 2} .6554742{col 62}{space 1}    0.52{col 71}{space 3}0.603{col 79}{space 4}-.9441804{col 92}{space 3} 1.625231
{txt}{space 22}newprojects_pct {c |}{col 39}{res}{space 2}-.7722883{col 51}{space 2} .1771817{col 62}{space 1}   -4.36{col 71}{space 3}0.000{col 79}{space 4}-1.119558{col 92}{space 3}-.4250186
{txt}{space 27}cio_tenure {c |}{col 39}{res}{space 2}-.0195397{col 51}{space 2} .0280677{col 62}{space 1}   -0.70{col 71}{space 3}0.486{col 79}{space 4}-.0745514{col 92}{space 3}  .035472
{txt}{space 35}y1 {c |}{col 39}{res}{space 2}-1.235054{col 51}{space 2} .1015185{col 62}{space 1}  -12.17{col 71}{space 3}0.000{col 79}{space 4}-1.434027{col 92}{space 3}-1.036082
{txt}{space 35}y2 {c |}{col 39}{res}{space 2}-1.186444{col 51}{space 2} .1218941{col 62}{space 1}   -9.73{col 71}{space 3}0.000{col 79}{space 4}-1.425352{col 92}{space 3}-.9475358
{txt}{space 35}y3 {c |}{col 39}{res}{space 2} -1.13669{col 51}{space 2} .1261777{col 62}{space 1}   -9.01{col 71}{space 3}0.000{col 79}{space 4}-1.383994{col 92}{space 3}-.8893862
{txt}{space 35}y4 {c |}{col 39}{res}{space 2}-.9237792{col 51}{space 2} .1347359{col 62}{space 1}   -6.86{col 71}{space 3}0.000{col 79}{space 4}-1.187857{col 92}{space 3}-.6597017
{txt}{space 35}y5 {c |}{col 39}{res}{space 2}-.9978891{col 51}{space 2} .1166018{col 62}{space 1}   -8.56{col 71}{space 3}0.000{col 79}{space 4}-1.226424{col 92}{space 3}-.7693538
{txt}{space 35}y6 {c |}{col 39}{res}{space 2}-1.003352{col 51}{space 2} .1112587{col 62}{space 1}   -9.02{col 71}{space 3}0.000{col 79}{space 4}-1.221415{col 92}{space 3}-.7852887
{txt}{space 35}y7 {c |}{col 39}{res}{space 2}        0{col 51}{txt}  (omitted)
{space 25}involuntaryb {c |}{col 39}{res}{space 2} 8.620299{col 51}{space 2} 5.368232{col 62}{space 1}    1.61{col 71}{space 3}0.108{col 79}{space 4}-1.901242{col 92}{space 3} 19.14184
{txt}{space 15}involabove15_los15pctb {c |}{col 39}{res}{space 2}-7.969282{col 51}{space 2} 31.45524{col 62}{space 1}   -0.25{col 71}{space 3}0.800{col 79}{space 4}-69.62043{col 92}{space 3} 53.68186
{txt}{space 15}activityno_permil_pctb {c |}{col 39}{res}{space 2} .8041256{col 51}{space 2} .3976253{col 62}{space 1}    2.02{col 71}{space 3}0.043{col 79}{space 4} .0247943{col 92}{space 3} 1.583457
{txt}{space 29}lnitempb {c |}{col 39}{res}{space 2} .4773138{col 51}{space 2} .3131809{col 62}{space 1}    1.52{col 71}{space 3}0.127{col 79}{space 4}-.1365095{col 92}{space 3} 1.091137
{txt}{space 26}lnprojsizeb {c |}{col 39}{res}{space 2} .0120985{col 51}{space 2} .0695769{col 62}{space 1}    0.17{col 71}{space 3}0.862{col 79}{space 4}-.1242698{col 92}{space 3} .1484667
{txt}{space 25}projects_nob {c |}{col 39}{res}{space 2}-.0038287{col 51}{space 2}  .002701{col 62}{space 1}   -1.42{col 71}{space 3}0.156{col 79}{space 4}-.0091225{col 92}{space 3} .0014651
{txt}{space 29}accrateb {c |}{col 39}{res}{space 2}-2.009592{col 51}{space 2} 1.469948{col 62}{space 1}   -1.37{col 71}{space 3}0.172{col 79}{space 4}-4.890638{col 92}{space 3} .8714539
{txt}{space 21}newprojects_pctb {c |}{col 39}{res}{space 2}-1.094074{col 51}{space 2} .3762056{col 62}{space 1}   -2.91{col 71}{space 3}0.004{col 79}{space 4}-1.831423{col 92}{space 3}-.3567246
{txt}{space 26}cio_tenureb {c |}{col 39}{res}{space 2}-.0120249{col 51}{space 2} .0629642{col 62}{space 1}   -0.19{col 71}{space 3}0.849{col 79}{space 4}-.1354325{col 92}{space 3} .1113827
{txt}{space 15}dmeratio_totalspending {c |}{col 39}{res}{space 2} -.098583{col 51}{space 2} .4010583{col 62}{space 1}   -0.25{col 71}{space 3}0.806{col 79}{space 4}-.8846428{col 92}{space 3} .6874768
{txt}{space 14}dmeratio_totalspendingb {c |}{col 39}{res}{space 2} .3880453{col 51}{space 2} .5071896{col 62}{space 1}    0.77{col 71}{space 3}0.444{col 79}{space 4}-.6060281{col 92}{space 3} 1.382119
{txt}{space 32}_cons {c |}{col 39}{res}{space 2} .8396074{col 51}{space 2} .4230218{col 62}{space 1}    1.98{col 71}{space 3}0.047{col 79}{space 4} .0104999{col 92}{space 3} 1.668715
{txt}{hline 38}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est6{txt} stored)

{com}. 
. * Aggregate Involuntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins , dydx(involuntary) at(involabove15_los15pct=($invollos_p10 $invollos_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.01}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} .6096748{col 26}{space 2} .4223076{col 37}{space 1}    1.44{col 46}{space 3}0.149{col 54}{space 4} -.218033{col 67}{space 3} 1.437383
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 4C]
. margins , dydx(involuntary) at(involabove15_los15pct=(0(0.01)0.05))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.01}}
{lalign 7:3._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.02}}
{lalign 7:4._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.03}}
{lalign 7:5._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.04}}
{lalign 7:6._at: }{space 0}{lalign 16:involabove15_l~t} = {res:{ralign 3:.05}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2}-1.652312{col 26}{space 2} .9816132{col 37}{space 1}   -1.68{col 46}{space 3}0.092{col 54}{space 4}-3.576239{col 67}{space 3} .2716141
{txt}{space 10}2  {c |}{col 14}{res}{space 2}-1.042638{col 26}{space 2} .9554041{col 37}{space 1}   -1.09{col 46}{space 3}0.275{col 54}{space 4}-2.915195{col 67}{space 3}   .82992
{txt}{space 10}3  {c |}{col 14}{res}{space 2}-.4803263{col 26}{space 2} 1.012631{col 37}{space 1}   -0.47{col 46}{space 3}0.635{col 54}{space 4}-2.465047{col 67}{space 3} 1.504394
{txt}{space 10}4  {c |}{col 14}{res}{space 2}  .029576{col 26}{space 2} 1.038508{col 37}{space 1}    0.03{col 46}{space 3}0.977{col 54}{space 4}-2.005862{col 67}{space 3} 2.065014
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .4835551{col 26}{space 2} .9753286{col 37}{space 1}    0.50{col 46}{space 3}0.620{col 54}{space 4}-1.428054{col 67}{space 3} 2.395164
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .8795785{col 26}{space 2} .8121343{col 37}{space 1}    1.08{col 46}{space 3}0.279{col 54}{space 4}-.7121755{col 67}{space 3} 2.471332
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// 
> xlabel(0 "0" 0.01 "1" 0.02 "2" 0.03 "3" 0.04 "4" 0.05 "5",format(%9.0f)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departures' Service Length (>15 years, %)", margin(t=1)  size(11pt)) ///
> title("{c -(}bf: Figure 4C. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Aggregate Involuntary Turnover Model [MODEL 14]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram involabove15_los15pct if involabove15_los15pct<=0.05, yaxis(2) xaxis(1)  fcolor(gray%20) yscale(axis(2) alt) lc(gray%10) xlabel(,format(%9.0f)) ylabel( ,format(%9.1f) angle(0) ) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:involabove15_los15pct}{p_end}
{res}{txt}
{com}. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4C.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4C.04-29-2025.gph} saved

{com}. 
. 
. 
. ***************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 15, FIGURE 5C] H3: Turnover X Agency Centralization
. eststo: xtgee delay_pct c.involuntary##i.cio5 c.involabove15_los15pct c.activityno_permil_pct   lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 involuntaryb involabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb  , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}3.2553936
{txt}Iteration 2:  Tolerance = {res}.12392093
{txt}Iteration 3:  Tolerance = {res}.30192453
{txt}Iteration 4:  Tolerance = {res}.10435214
{txt}Iteration 5:  Tolerance = {res}.03492712
{txt}Iteration 6:  Tolerance = {res}.01194845
{txt}Iteration 7:  Tolerance = {res}.00355374
{txt}Iteration 8:  Tolerance = {res}.00118163
{txt}Iteration 9:  Tolerance = {res}.00078198
{txt}Iteration 10: Tolerance = {res}.0004753
{txt}Iteration 11: Tolerance = {res}.00028114
{txt}Iteration 12: Tolerance = {res}.00018824
{txt}Iteration 13: Tolerance = {res}.00011511
{txt}Iteration 14: Tolerance = {res}.00006672
{txt}Iteration 15: Tolerance = {res}.00003734
{txt}Iteration 16: Tolerance = {res}.00002038
{txt}Iteration 17: Tolerance = {res}.00001092
{txt}Iteration 18: Tolerance = {res}5.763e-06
{txt}Iteration 19: Tolerance = {res}3.008e-06
{txt}Iteration 20: Tolerance = {res}1.555e-06
{txt}Iteration 21: Tolerance = {res}7.974e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:340.27}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 12}involuntary {c |}{col 25}{res}{space 2}-2.597697{col 37}{space 2} 3.397523{col 48}{space 1}   -0.76{col 57}{space 3}0.445{col 65}{space 4} -9.25672{col 78}{space 3} 4.061325
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.0494026{col 37}{space 2} .1011573{col 48}{space 1}   -0.49{col 57}{space 3}0.625{col 65}{space 4}-.2476674{col 78}{space 3} .1488621
{txt}{space 23} {c |}
{space 5}cio5#c.involuntary {c |}
{space 21}1  {c |}{col 25}{res}{space 2}-2.943692{col 37}{space 2} 3.002038{col 48}{space 1}   -0.98{col 57}{space 3}0.327{col 65}{space 4}-8.827578{col 78}{space 3} 2.940193
{txt}{space 23} {c |}
{space 2}involabove15_los15pct {c |}{col 25}{res}{space 2}-.9272482{col 37}{space 2} 12.50983{col 48}{space 1}   -0.07{col 57}{space 3}0.941{col 65}{space 4}-25.44607{col 78}{space 3} 23.59157
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2441537{col 37}{space 2} .2233769{col 48}{space 1}    1.09{col 57}{space 3}0.274{col 65}{space 4}-.1936571{col 78}{space 3} .6819644
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2}-.4049505{col 37}{space 2} .3054406{col 48}{space 1}   -1.33{col 57}{space 3}0.185{col 65}{space 4}-1.003603{col 78}{space 3} .1937021
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2}    .0303{col 37}{space 2} .0301811{col 48}{space 1}    1.00{col 57}{space 3}0.315{col 65}{space 4}-.0288539{col 78}{space 3} .0894539
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0012884{col 37}{space 2} .0016706{col 48}{space 1}    0.77{col 57}{space 3}0.441{col 65}{space 4}-.0019859{col 78}{space 3} .0045627
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .3574225{col 37}{space 2} .6459729{col 48}{space 1}    0.55{col 57}{space 3}0.580{col 65}{space 4}-.9086612{col 78}{space 3} 1.623506
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.7870155{col 37}{space 2} .1757563{col 48}{space 1}   -4.48{col 57}{space 3}0.000{col 65}{space 4}-1.131492{col 78}{space 3}-.4425396
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0161759{col 37}{space 2}  .027628{col 48}{space 1}   -0.59{col 57}{space 3}0.558{col 65}{space 4}-.0703257{col 78}{space 3} .0379739
{txt}{space 21}y1 {c |}{col 25}{res}{space 2} -1.23366{col 37}{space 2} .1020083{col 48}{space 1}  -12.09{col 57}{space 3}0.000{col 65}{space 4}-1.433592{col 78}{space 3}-1.033727
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.187188{col 37}{space 2} .1226248{col 48}{space 1}   -9.68{col 57}{space 3}0.000{col 65}{space 4}-1.427528{col 78}{space 3}-.9468474
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.140409{col 37}{space 2} .1259282{col 48}{space 1}   -9.06{col 57}{space 3}0.000{col 65}{space 4}-1.387224{col 78}{space 3}-.8935941
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.9140411{col 37}{space 2} .1313068{col 48}{space 1}   -6.96{col 57}{space 3}0.000{col 65}{space 4}-1.171398{col 78}{space 3}-.6566845
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9936305{col 37}{space 2} .1164075{col 48}{space 1}   -8.54{col 57}{space 3}0.000{col 65}{space 4}-1.221785{col 78}{space 3} -.765476
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-1.003159{col 37}{space 2}   .11263{col 48}{space 1}   -8.91{col 57}{space 3}0.000{col 65}{space 4} -1.22391{col 78}{space 3}-.7824085
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 11}involuntaryb {c |}{col 25}{res}{space 2} 7.718944{col 37}{space 2} 5.364991{col 48}{space 1}    1.44{col 57}{space 3}0.150{col 65}{space 4}-2.796246{col 78}{space 3} 18.23413
{txt}{space 1}involabove15_los15pctb {c |}{col 25}{res}{space 2}-16.32344{col 37}{space 2} 29.30393{col 48}{space 1}   -0.56{col 57}{space 3}0.578{col 65}{space 4}-73.75809{col 78}{space 3} 41.11121
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8014092{col 37}{space 2} .3943922{col 48}{space 1}    2.03{col 57}{space 3}0.042{col 65}{space 4} .0284148{col 78}{space 3} 1.574404
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} .4436633{col 37}{space 2}  .315316{col 48}{space 1}    1.41{col 57}{space 3}0.159{col 65}{space 4}-.1743448{col 78}{space 3} 1.061671
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0087735{col 37}{space 2} .0697346{col 48}{space 1}    0.13{col 57}{space 3}0.900{col 65}{space 4}-.1279038{col 78}{space 3} .1454508
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0038737{col 37}{space 2} .0026772{col 48}{space 1}   -1.45{col 57}{space 3}0.148{col 65}{space 4}-.0091209{col 78}{space 3} .0013736
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-2.039789{col 37}{space 2} 1.461277{col 48}{space 1}   -1.40{col 57}{space 3}0.163{col 65}{space 4}-4.903839{col 78}{space 3} .8242617
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.073366{col 37}{space 2} .3750769{col 48}{space 1}   -2.86{col 57}{space 3}0.004{col 65}{space 4}-1.808503{col 78}{space 3}-.3382289
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0152103{col 37}{space 2} .0629295{col 48}{space 1}   -0.24{col 57}{space 3}0.809{col 65}{space 4}-.1385498{col 78}{space 3} .1081293
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2}-.0515382{col 37}{space 2}  .397342{col 48}{space 1}   -0.13{col 57}{space 3}0.897{col 65}{space 4}-.8303142{col 78}{space 3} .7272378
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3600013{col 37}{space 2} .5053658{col 48}{space 1}    0.71{col 57}{space 3}0.476{col 65}{space 4}-.6304975{col 78}{space 3}   1.3505
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .8224501{col 37}{space 2} .4223027{col 48}{space 1}    1.95{col 57}{space 3}0.051{col 65}{space 4}-.0052479{col 78}{space 3} 1.650148
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est7{txt} stored)

{com}. 
. * Aggregate Involuntary Turnover Rate: Marginal Effect Differentials Based on Agency Centralization
. margins , dydx(involuntary) over(cio5) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 8}cio5 {c |}
{space 5}1 vs 0  {c |}{col 14}{res}{space 2}-.8749078{col 26}{space 2} .9276536{col 37}{space 1}   -0.94{col 46}{space 3}0.346{col 54}{space 4}-2.693075{col 67}{space 3} .9432598
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. 
. * [FIGURE 5C]
. margins , dydx(involuntary) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 8}cio5 {c |}
{space 10}0  {c |}{col 14}{res}{space 2}-.7875949{col 26}{space 2} 1.030328{col 37}{space 1}   -0.76{col 46}{space 3}0.445{col 54}{space 4}-2.807002{col 67}{space 3} 1.231812
{txt}{space 10}1  {c |}{col 14}{res}{space 2}-1.662503{col 26}{space 2}  1.07571{col 37}{space 1}   -1.55{col 46}{space 3}0.122{col 54}{space 4}-3.770856{col 67}{space 3} .4458511
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(black)) ciopt(color(black%60))  yline(0, lcolor(red) lpattern(shortdash)) ///
> xlabel (-1 " " 0 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel( ,format(%9.1f) angle(0)) ysc(r(-4(1)2)) ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5C. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Aggregate Involuntary Turnover Model [MODEL 15]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5C.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5C.04-29-2025.gph} saved

{com}. 
. 
. 
. ***************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 16, FIGURE 6C] H4: Turnover X Task Standardization
. eststo: xtgee delay_pct c.involuntary##c.activityno_permil_pct c.involabove15_los15pct  i.cio5 lnitemp lnprojsize projects_no accrate  newprojects_pct  cio_tenure y1-y7 involuntaryb involabove15_los15pctb  activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}3.9675312
{txt}Iteration 2:  Tolerance = {res}.79507137
{txt}Iteration 3:  Tolerance = {res}.87684065
{txt}Iteration 4:  Tolerance = {res}.0990977
{txt}Iteration 5:  Tolerance = {res}.03072806
{txt}Iteration 6:  Tolerance = {res}.01683786
{txt}Iteration 7:  Tolerance = {res}.01461824
{txt}Iteration 8:  Tolerance = {res}.0098135
{txt}Iteration 9:  Tolerance = {res}.00590036
{txt}Iteration 10: Tolerance = {res}.00333036
{txt}Iteration 11: Tolerance = {res}.00180305
{txt}Iteration 12: Tolerance = {res}.0009472
{txt}Iteration 13: Tolerance = {res}.0004861
{txt}Iteration 14: Tolerance = {res}.00024472
{txt}Iteration 15: Tolerance = {res}.00012118
{txt}Iteration 16: Tolerance = {res}.00005912
{txt}Iteration 17: Tolerance = {res}.00002845
{txt}Iteration 18: Tolerance = {res}.00001351
{txt}Iteration 19: Tolerance = {res}6.333e-06
{txt}Iteration 20: Tolerance = {res}2.930e-06
{txt}Iteration 21: Tolerance = {res}1.337e-06
{txt}Iteration 22: Tolerance = {res}6.012e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:540}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:94}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:335.22}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 103:(Std. err. adjusted for clustering on {res:a_id})}
{hline 38}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 39}{c |}{col 51}  Semirobust
{col 1}                            delay_pct{col 39}{c |} Coefficient{col 51}  std. err.{col 63}      z{col 71}   P>|z|{col 79}     [95% con{col 92}f. interval]
{hline 38}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}involuntary {c |}{col 39}{res}{space 2}-3.021576{col 51}{space 2} 4.376794{col 62}{space 1}   -0.69{col 71}{space 3}0.490{col 79}{space 4}-11.59994{col 92}{space 3} 5.556783
{txt}{space 16}activityno_permil_pct {c |}{col 39}{res}{space 2} .2437065{col 51}{space 2} .2297758{col 62}{space 1}    1.06{col 71}{space 3}0.289{col 79}{space 4}-.2066459{col 92}{space 3} .6940588
{txt}{space 37} {c |}
c.involuntary#c.activityno_permil_pct {c |}{col 39}{res}{space 2}-.9352962{col 51}{space 2} 10.02352{col 62}{space 1}   -0.09{col 71}{space 3}0.926{col 79}{space 4}-20.58103{col 92}{space 3} 18.71044
{txt}{space 37} {c |}
{space 16}involabove15_los15pct {c |}{col 39}{res}{space 2} 2.408106{col 51}{space 2} 11.71761{col 62}{space 1}    0.21{col 71}{space 3}0.837{col 79}{space 4}-20.55798{col 92}{space 3}  25.3742
{txt}{space 31}1.cio5 {c |}{col 39}{res}{space 2} -.061283{col 51}{space 2} .0982638{col 62}{space 1}   -0.62{col 71}{space 3}0.533{col 79}{space 4}-.2538764{col 92}{space 3} .1313105
{txt}{space 30}lnitemp {c |}{col 39}{res}{space 2}-.4043166{col 51}{space 2} .3076768{col 62}{space 1}   -1.31{col 71}{space 3}0.189{col 79}{space 4}-1.007352{col 92}{space 3} .1987189
{txt}{space 27}lnprojsize {c |}{col 39}{res}{space 2} .0304752{col 51}{space 2} .0318768{col 62}{space 1}    0.96{col 71}{space 3}0.339{col 79}{space 4}-.0320021{col 92}{space 3} .0929525
{txt}{space 26}projects_no {c |}{col 39}{res}{space 2} .0012872{col 51}{space 2} .0016771{col 62}{space 1}    0.77{col 71}{space 3}0.443{col 79}{space 4}-.0019999{col 92}{space 3} .0045743
{txt}{space 30}accrate {c |}{col 39}{res}{space 2} .3832708{col 51}{space 2} .6322932{col 62}{space 1}    0.61{col 71}{space 3}0.544{col 79}{space 4}-.8560011{col 92}{space 3} 1.622543
{txt}{space 22}newprojects_pct {c |}{col 39}{res}{space 2}-.7897526{col 51}{space 2} .1742516{col 62}{space 1}   -4.53{col 71}{space 3}0.000{col 79}{space 4} -1.13128{col 92}{space 3}-.4482257
{txt}{space 27}cio_tenure {c |}{col 39}{res}{space 2}-.0150488{col 51}{space 2} .0275289{col 62}{space 1}   -0.55{col 71}{space 3}0.585{col 79}{space 4}-.0690045{col 92}{space 3}  .038907
{txt}{space 35}y1 {c |}{col 39}{res}{space 2}-1.231384{col 51}{space 2} .1024196{col 62}{space 1}  -12.02{col 71}{space 3}0.000{col 79}{space 4}-1.432123{col 92}{space 3}-1.030645
{txt}{space 35}y2 {c |}{col 39}{res}{space 2}-1.185599{col 51}{space 2} .1230162{col 62}{space 1}   -9.64{col 71}{space 3}0.000{col 79}{space 4}-1.426706{col 92}{space 3}-.9444919
{txt}{space 35}y3 {c |}{col 39}{res}{space 2}-1.139726{col 51}{space 2} .1256007{col 62}{space 1}   -9.07{col 71}{space 3}0.000{col 79}{space 4}-1.385899{col 92}{space 3}-.8935532
{txt}{space 35}y4 {c |}{col 39}{res}{space 2}-.9078612{col 51}{space 2} .1294171{col 62}{space 1}   -7.01{col 71}{space 3}0.000{col 79}{space 4}-1.161514{col 92}{space 3}-.6542083
{txt}{space 35}y5 {c |}{col 39}{res}{space 2} -.992057{col 51}{space 2} .1163199{col 62}{space 1}   -8.53{col 71}{space 3}0.000{col 79}{space 4} -1.22004{col 92}{space 3}-.7640742
{txt}{space 35}y6 {c |}{col 39}{res}{space 2}-1.000705{col 51}{space 2} .1125182{col 62}{space 1}   -8.89{col 71}{space 3}0.000{col 79}{space 4}-1.221237{col 92}{space 3}-.7801739
{txt}{space 35}y7 {c |}{col 39}{res}{space 2}        0{col 51}{txt}  (omitted)
{space 25}involuntaryb {c |}{col 39}{res}{space 2} 7.511109{col 51}{space 2} 5.580424{col 62}{space 1}    1.35{col 71}{space 3}0.178{col 79}{space 4}-3.426322{col 92}{space 3} 18.44854
{txt}{space 15}involabove15_los15pctb {c |}{col 39}{res}{space 2}-19.31042{col 51}{space 2} 29.32661{col 62}{space 1}   -0.66{col 71}{space 3}0.510{col 79}{space 4}-76.78952{col 92}{space 3} 38.16868
{txt}{space 15}activityno_permil_pctb {c |}{col 39}{res}{space 2} .8032607{col 51}{space 2} .3935093{col 62}{space 1}    2.04{col 71}{space 3}0.041{col 79}{space 4} .0319967{col 92}{space 3} 1.574525
{txt}{space 29}lnitempb {c |}{col 39}{res}{space 2} .4441276{col 51}{space 2}  .317875{col 62}{space 1}    1.40{col 71}{space 3}0.162{col 79}{space 4}-.1788959{col 92}{space 3} 1.067151
{txt}{space 26}lnprojsizeb {c |}{col 39}{res}{space 2} .0073103{col 51}{space 2} .0701953{col 62}{space 1}    0.10{col 71}{space 3}0.917{col 79}{space 4}  -.13027{col 92}{space 3} .1448907
{txt}{space 25}projects_nob {c |}{col 39}{res}{space 2}-.0038335{col 51}{space 2} .0026839{col 62}{space 1}   -1.43{col 71}{space 3}0.153{col 79}{space 4}-.0090939{col 92}{space 3} .0014268
{txt}{space 29}accrateb {c |}{col 39}{res}{space 2}-2.055896{col 51}{space 2} 1.447474{col 62}{space 1}   -1.42{col 71}{space 3}0.156{col 79}{space 4}-4.892894{col 92}{space 3} .7811017
{txt}{space 21}newprojects_pctb {c |}{col 39}{res}{space 2}-1.068786{col 51}{space 2} .3744957{col 62}{space 1}   -2.85{col 71}{space 3}0.004{col 79}{space 4}-1.802784{col 92}{space 3}-.3347875
{txt}{space 26}cio_tenureb {c |}{col 39}{res}{space 2}-.0180695{col 51}{space 2} .0633629{col 62}{space 1}   -0.29{col 71}{space 3}0.776{col 79}{space 4}-.1422585{col 92}{space 3} .1061194
{txt}{space 15}dmeratio_totalspending {c |}{col 39}{res}{space 2}-.0429536{col 51}{space 2} .3956831{col 62}{space 1}   -0.11{col 71}{space 3}0.914{col 79}{space 4}-.8184783{col 92}{space 3}  .732571
{txt}{space 14}dmeratio_totalspendingb {c |}{col 39}{res}{space 2} .3601893{col 51}{space 2} .5034924{col 62}{space 1}    0.72{col 71}{space 3}0.474{col 79}{space 4}-.6266376{col 92}{space 3} 1.347016
{txt}{space 32}_cons {c |}{col 39}{res}{space 2} .8199918{col 51}{space 2}  .425584{col 62}{space 1}    1.93{col 71}{space 3}0.054{col 79}{space 4}-.0141375{col 92}{space 3} 1.654121
{txt}{hline 38}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est8{txt} stored)

{com}. 
. * Aggregate Involuntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Task Standardization
. margins , dydx(involuntary) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}   Contrast{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2}-.2880078{col 26}{space 2} 2.190852{col 37}{space 1}   -0.13{col 46}{space 3}0.895{col 54}{space 4}-4.581999{col 67}{space 3} 4.005984
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. 
. * [FIGURE 6C]
. margins , dydx(involuntary) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:540}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}      dy/dx{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary  {txt}{c |}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2}  -.88234{col 26}{space 2}  1.28387{col 37}{space 1}   -0.69{col 46}{space 3}0.492{col 54}{space 4}-3.398679{col 67}{space 3} 1.633999
{txt}{space 10}2  {c |}{col 14}{res}{space 2} -.921696{col 26}{space 2} 1.098406{col 37}{space 1}   -0.84{col 46}{space 3}0.401{col 54}{space 4}-3.074531{col 67}{space 3} 1.231139
{txt}{space 10}3  {c |}{col 14}{res}{space 2}-.9615161{col 26}{space 2}  .960193{col 37}{space 1}   -1.00{col 46}{space 3}0.317{col 54}{space 4} -2.84346{col 67}{space 3} .9204276
{txt}{space 10}4  {c |}{col 14}{res}{space 2}-1.001762{col 26}{space 2} .8992566{col 37}{space 1}   -1.11{col 46}{space 3}0.265{col 54}{space 4}-2.764273{col 67}{space 3} .7607486
{txt}{space 10}5  {c |}{col 14}{res}{space 2}-1.042394{col 26}{space 2} .9381944{col 37}{space 1}   -1.11{col 46}{space 3}0.267{col 54}{space 4}-2.881221{col 67}{space 3} .7964335
{txt}{space 10}6  {c |}{col 14}{res}{space 2} -1.08337{col 26}{space 2} 1.072321{col 37}{space 1}   -1.01{col 46}{space 3}0.312{col 54}{space 4} -3.18508{col 67}{space 3} 1.018341
{txt}{space 10}7  {c |}{col 14}{res}{space 2}-1.124648{col 26}{space 2} 1.276886{col 37}{space 1}   -0.88{col 46}{space 3}0.378{col 54}{space 4}-3.627298{col 67}{space 3} 1.378003
{txt}{space 10}8  {c |}{col 14}{res}{space 2}-1.166182{col 26}{space 2} 1.527778{col 37}{space 1}   -0.76{col 46}{space 3}0.445{col 54}{space 4}-4.160572{col 67}{space 3} 1.828208
{txt}{space 10}9  {c |}{col 14}{res}{space 2}-1.207928{col 26}{space 2} 1.808977{col 37}{space 1}   -0.67{col 46}{space 3}0.504{col 54}{space 4}-4.753457{col 67}{space 3} 2.337602
{txt}{space 9}10  {c |}{col 14}{res}{space 2}-1.249837{col 26}{space 2} 2.110934{col 37}{space 1}   -0.59{col 46}{space 3}0.554{col 54}{space 4}-5.387191{col 67}{space 3} 2.887517
{txt}{space 9}11  {c |}{col 14}{res}{space 2}-1.291862{col 26}{space 2} 2.427939{col 37}{space 1}   -0.53{col 46}{space 3}0.595{col 54}{space 4}-6.050535{col 67}{space 3}  3.46681
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100", format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", size(11pt) margin(t=1)) ///
> title("{c -(}bf: Figure 6C. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Aggregate Involuntary Turnover Model [MODEL 16]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100", format(%9.0f)) ylabel(,format(%9.1f) angle(0))  ylabel( ,format(%9.1f) angle(0) axis(2)) ytitle(,  axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6C.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6C.04-29-2025.gph} saved

{com}. 
. *
. ****************************************************************************************************************************************************
. ** [TABLE A2. MODELS A9 - A16; AGGREGATE VOLUNTARY AND INVOLUNTARY TURNOVER MODELS]
. 
. cd "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES"
{res}C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES
{txt}
{com}. 
. esttab est1 est2 est3 est4 using ManuscriptTable2.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 2. Fractional Panel Probit Analysis of the Impact of Aggregate Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 9" "Model 10" "Model 11" "Model 12" )  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) replace
{res}{txt}(output written to {browse  `"ManuscriptTable2.04-29-2025.rtf"'})

{com}. 
. esttab est5 est6 est7 est8 using ManuscriptTable2.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 2. Fractional Panel Probit Analysis of the Impact of Aggregate Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 13" "Model 14" "Model 15" "Model 16")  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) append
{res}{txt}(output written to {browse  `"ManuscriptTable2.04-29-2025.rtf"'})

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. **# [MODELS 17 - 20, FIGURES 3E - 6E] MANAGERIAL VOLUNTARY TURNOVER MODELS
. 
. discard
{txt}
{com}. 
. quietly: xtgee delay_pct c.turnover_gs13  gs13above15_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnover_gs13b gs13above15_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}
{com}. 
. * [MODEL 17, FIGURE 3E] H1: Managerial Voluntary Turnover Baseline Effects
. eststo: xtgee delay_pct c.voluntary_gs13 c.volgs13_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.voluntary_gs13b c.volgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.35235176
{txt}Iteration 2:  Tolerance = {res}.18591623
{txt}Iteration 3:  Tolerance = {res}.07031065
{txt}Iteration 4:  Tolerance = {res}.02619746
{txt}Iteration 5:  Tolerance = {res}.00894424
{txt}Iteration 6:  Tolerance = {res}.00354473
{txt}Iteration 7:  Tolerance = {res}.0015783
{txt}Iteration 8:  Tolerance = {res}.00069202
{txt}Iteration 9:  Tolerance = {res}.00030139
{txt}Iteration 10: Tolerance = {res}.00013089
{txt}Iteration 11: Tolerance = {res}.00005678
{txt}Iteration 12: Tolerance = {res}.00002462
{txt}Iteration 13: Tolerance = {res}.00001067
{txt}Iteration 14: Tolerance = {res}4.627e-06
{txt}Iteration 15: Tolerance = {res}2.006e-06
{txt}Iteration 16: Tolerance = {res}8.696e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:326.97}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}voluntary_gs13 {c |}{col 25}{res}{space 2}  .460395{col 37}{space 2} 1.885236{col 48}{space 1}    0.24{col 57}{space 3}0.807{col 65}{space 4}  -3.2346{col 78}{space 3}  4.15539
{txt}{space 7}volgs13_los15pct {c |}{col 25}{res}{space 2}-.2330909{col 37}{space 2} 1.102087{col 48}{space 1}   -0.21{col 57}{space 3}0.832{col 65}{space 4}-2.393141{col 78}{space 3} 1.926959
{txt}{space 19}cio5 {c |}{col 25}{res}{space 2}-.0915661{col 37}{space 2} .0823593{col 48}{space 1}   -1.11{col 57}{space 3}0.266{col 65}{space 4}-.2529875{col 78}{space 3} .0698552
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2793034{col 37}{space 2} .2003338{col 48}{space 1}    1.39{col 57}{space 3}0.163{col 65}{space 4}-.1133436{col 78}{space 3} .6719505
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0125225{col 37}{space 2} .0243824{col 48}{space 1}   -0.51{col 57}{space 3}0.608{col 65}{space 4}-.0603111{col 78}{space 3} .0352661
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .1158071{col 37}{space 2} .2368848{col 48}{space 1}    0.49{col 57}{space 3}0.625{col 65}{space 4}-.3484785{col 78}{space 3} .5800927
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2}  .078729{col 37}{space 2} .0295066{col 48}{space 1}    2.67{col 57}{space 3}0.008{col 65}{space 4} .0208972{col 78}{space 3} .1365608
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0011025{col 37}{space 2} .0017555{col 48}{space 1}    0.63{col 57}{space 3}0.530{col 65}{space 4}-.0023382{col 78}{space 3} .0045432
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .4011976{col 37}{space 2} .5463821{col 48}{space 1}    0.73{col 57}{space 3}0.463{col 65}{space 4}-.6696916{col 78}{space 3} 1.472087
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8163754{col 37}{space 2} .1658251{col 48}{space 1}   -4.92{col 57}{space 3}0.000{col 65}{space 4}-1.141387{col 78}{space 3}-.4913642
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.191975{col 37}{space 2} .1042294{col 48}{space 1}  -11.44{col 57}{space 3}0.000{col 65}{space 4}-1.396261{col 78}{space 3}-.9876896
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.177256{col 37}{space 2}  .126231{col 48}{space 1}   -9.33{col 57}{space 3}0.000{col 65}{space 4}-1.424665{col 78}{space 3} -.929848
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.119537{col 37}{space 2} .1180104{col 48}{space 1}   -9.49{col 57}{space 3}0.000{col 65}{space 4}-1.350833{col 78}{space 3}-.8882409
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.8678378{col 37}{space 2}  .119867{col 48}{space 1}   -7.24{col 57}{space 3}0.000{col 65}{space 4}-1.102773{col 78}{space 3}-.6329027
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9996571{col 37}{space 2} .1098391{col 48}{space 1}   -9.10{col 57}{space 3}0.000{col 65}{space 4}-1.214938{col 78}{space 3}-.7843765
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8992018{col 37}{space 2} .1031011{col 48}{space 1}   -8.72{col 57}{space 3}0.000{col 65}{space 4}-1.101276{col 78}{space 3}-.6971272
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 8}voluntary_gs13b {c |}{col 25}{res}{space 2}-.2223988{col 37}{space 2} 3.191549{col 48}{space 1}   -0.07{col 57}{space 3}0.944{col 65}{space 4}-6.477719{col 78}{space 3} 6.032922
{txt}{space 6}volgs13_los15pctb {c |}{col 25}{res}{space 2}-6.511787{col 37}{space 2} 2.290719{col 48}{space 1}   -2.84{col 57}{space 3}0.004{col 65}{space 4}-11.00151{col 78}{space 3}-2.022061
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9876362{col 37}{space 2} .4070265{col 48}{space 1}    2.43{col 57}{space 3}0.015{col 65}{space 4} .1898789{col 78}{space 3} 1.785394
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2}-.1097821{col 37}{space 2} .2422722{col 48}{space 1}   -0.45{col 57}{space 3}0.650{col 65}{space 4}-.5846269{col 78}{space 3} .3650626
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0020414{col 37}{space 2} .0640197{col 48}{space 1}    0.03{col 57}{space 3}0.975{col 65}{space 4} -.123435{col 78}{space 3} .1275178
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0054243{col 37}{space 2} .0032967{col 48}{space 1}   -1.65{col 57}{space 3}0.100{col 65}{space 4}-.0118858{col 78}{space 3} .0010371
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-.9171912{col 37}{space 2}  1.22956{col 48}{space 1}   -0.75{col 57}{space 3}0.456{col 65}{space 4}-3.327085{col 78}{space 3} 1.492702
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.012155{col 37}{space 2}  .338594{col 48}{space 1}   -2.99{col 57}{space 3}0.003{col 65}{space 4}-1.675787{col 78}{space 3}-.3485234
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0342182{col 37}{space 2} .0553667{col 48}{space 1}   -0.62{col 57}{space 3}0.537{col 65}{space 4}-.1427349{col 78}{space 3} .0742986
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .3188415{col 37}{space 2} .4285694{col 48}{space 1}    0.74{col 57}{space 3}0.457{col 65}{space 4}-.5211391{col 78}{space 3} 1.158822
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3764306{col 37}{space 2} .5423264{col 48}{space 1}    0.69{col 57}{space 3}0.488{col 65}{space 4}-.6865096{col 78}{space 3} 1.439371
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .7738971{col 37}{space 2} .3567338{col 48}{space 1}    2.17{col 57}{space 3}0.030{col 65}{space 4} .0747118{col 78}{space 3} 1.473082
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est1{txt} stored)

{com}. 
. * Managerial Voluntary Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(voluntary_gs13)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2colreset}{...}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}{col 28} Delta-method
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
voluntary_gs13 {c |}{col 16}{res}{space 2} .1378493{col 28}{space 2} .5649484{col 39}{space 1}    0.24{col 48}{space 3}0.807{col 56}{space 4}-.9694293{col 69}{space 3} 1.245128
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Managerial Voluntary Turnover Rate
. margins, at(voluntary_gs13=($voluntary_gs13_p10 $voluntary_gs13_p90)) pwcompare(effects)
{res}
{txt}{col 1}Pairwise comparisons of predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.07}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method{col 37}    Una{col 46}djusted{col 54}          Una{col 67}djusted
{col 14}{c |}   Contrast{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 5}2 vs 1  {c |}{col 14}{res}{space 2} .0096599{col 26}{space 2} .0396306{col 37}{space 1}    0.24{col 46}{space 3}0.807{col 54}{space 4}-.0680146{col 67}{space 3} .0873344
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3E]
. margins, at(voluntary_gs13=(0(0.01)0.15)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:0}}
{lalign 8:2._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.01}}
{lalign 8:3._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.02}}
{lalign 8:4._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.03}}
{lalign 8:5._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.04}}
{lalign 8:6._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.05}}
{lalign 8:7._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.06}}
{lalign 8:8._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.07}}
{lalign 8:9._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.08}}
{lalign 8:10._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.09}}
{lalign 8:11._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.1}}
{lalign 8:12._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.11}}
{lalign 8:13._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.12}}
{lalign 8:14._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.13}}
{lalign 8:15._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.14}}
{lalign 8:16._at: }{space 0}{lalign 14:voluntary_gs13} = {res:{ralign 3:.15}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2878859{col 26}{space 2}  .020618{col 37}{space 1}   13.96{col 46}{space 3}0.000{col 54}{space 4} .2474753{col 67}{space 3} .3282965
{txt}{space 10}2  {c |}{col 14}{res}{space 2} .2892563{col 26}{space 2}  .017064{col 37}{space 1}   16.95{col 46}{space 3}0.000{col 54}{space 4} .2558115{col 67}{space 3} .3227011
{txt}{space 10}3  {c |}{col 14}{res}{space 2}   .29063{col 26}{space 2} .0148124{col 37}{space 1}   19.62{col 46}{space 3}0.000{col 54}{space 4} .2615981{col 67}{space 3} .3196618
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2920068{col 26}{space 2} .0145147{col 37}{space 1}   20.12{col 46}{space 3}0.000{col 54}{space 4} .2635586{col 67}{space 3}  .320455
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2933868{col 26}{space 2} .0163052{col 37}{space 1}   17.99{col 46}{space 3}0.000{col 54}{space 4} .2614292{col 67}{space 3} .3253444
{txt}{space 10}6  {c |}{col 14}{res}{space 2}   .29477{col 26}{space 2} .0196428{col 37}{space 1}   15.01{col 46}{space 3}0.000{col 54}{space 4} .2562709{col 67}{space 3} .3332691
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .2961563{col 26}{space 2} .0239057{col 37}{space 1}   12.39{col 46}{space 3}0.000{col 54}{space 4}  .249302{col 67}{space 3} .3430106
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .2975457{col 26}{space 2} .0286995{col 37}{space 1}   10.37{col 46}{space 3}0.000{col 54}{space 4} .2412958{col 67}{space 3} .3537957
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .2989383{col 26}{space 2} .0338117{col 37}{space 1}    8.84{col 46}{space 3}0.000{col 54}{space 4} .2326685{col 67}{space 3}  .365208
{txt}{space 9}10  {c |}{col 14}{res}{space 2} .3003339{col 26}{space 2} .0391287{col 37}{space 1}    7.68{col 46}{space 3}0.000{col 54}{space 4}  .223643{col 67}{space 3} .3770249
{txt}{space 9}11  {c |}{col 14}{res}{space 2} .3017327{col 26}{space 2} .0445869{col 37}{space 1}    6.77{col 46}{space 3}0.000{col 54}{space 4}  .214344{col 67}{space 3} .3891213
{txt}{space 9}12  {c |}{col 14}{res}{space 2} .3031344{col 26}{space 2} .0501485{col 37}{space 1}    6.04{col 46}{space 3}0.000{col 54}{space 4} .2048452{col 67}{space 3} .4014237
{txt}{space 9}13  {c |}{col 14}{res}{space 2} .3045393{col 26}{space 2} .0557902{col 37}{space 1}    5.46{col 46}{space 3}0.000{col 54}{space 4} .1951925{col 67}{space 3} .4138861
{txt}{space 9}14  {c |}{col 14}{res}{space 2} .3059472{col 26}{space 2} .0614967{col 37}{space 1}    4.98{col 46}{space 3}0.000{col 54}{space 4} .1854158{col 67}{space 3} .4264785
{txt}{space 9}15  {c |}{col 14}{res}{space 2}  .307358{col 26}{space 2} .0672578{col 37}{space 1}    4.57{col 46}{space 3}0.000{col 54}{space 4} .1755352{col 67}{space 3} .4391808
{txt}{space 9}16  {c |}{col 14}{res}{space 2} .3087719{col 26}{space 2}  .073066{col 37}{space 1}    4.23{col 46}{space 3}0.000{col 54}{space 4} .1655652{col 67}{space 3} .4519787
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) recastci(rarea) ///
> xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Managerial Voluntary Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3E. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Managerial Voluntary Turnover Model [MODEL 17]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram voluntary_gs13 if voluntary_gs13<=0.15, fc(gray%20) lc(gray%10) yaxis(2) yscale(axis(2) alt) xlabel(0 "0" 0.05 "5" 0.1 "10" 0.15 "15",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0)) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:voluntary_gs13}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3E.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3E.04-29-2025.gph} saved

{com}. 
. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 18, FIGURE 4E] H2: Turnover X Service Length
. eststo: xtgee delay_pct c.voluntary_gs13##c.volgs13_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.voluntary_gs13b##c.volgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.95988787
{txt}Iteration 2:  Tolerance = {res}.19498816
{txt}Iteration 3:  Tolerance = {res}.05939702
{txt}Iteration 4:  Tolerance = {res}.02035746
{txt}Iteration 5:  Tolerance = {res}.01064648
{txt}Iteration 6:  Tolerance = {res}.00553594
{txt}Iteration 7:  Tolerance = {res}.00282308
{txt}Iteration 8:  Tolerance = {res}.00141842
{txt}Iteration 9:  Tolerance = {res}.00070581
{txt}Iteration 10: Tolerance = {res}.00035584
{txt}Iteration 11: Tolerance = {res}.0001859
{txt}Iteration 12: Tolerance = {res}.00009499
{txt}Iteration 13: Tolerance = {res}.00004788
{txt}Iteration 14: Tolerance = {res}.00002392
{txt}Iteration 15: Tolerance = {res}.00001189
{txt}Iteration 16: Tolerance = {res}5.885e-06
{txt}Iteration 17: Tolerance = {res}2.907e-06
{txt}Iteration 18: Tolerance = {res}1.433e-06
{txt}Iteration 19: Tolerance = {res}7.061e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:29})} = {res}{ralign 6:355.19}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 103:(Std. err. adjusted for clustering on {res:a_id})}
{hline 38}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 39}{c |}{col 51}  Semirobust
{col 1}                            delay_pct{col 39}{c |} Coefficient{col 51}  std. err.{col 63}      z{col 71}   P>|z|{col 79}     [95% con{col 92}f. interval]
{hline 38}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 23}voluntary_gs13 {c |}{col 39}{res}{space 2} .2069915{col 51}{space 2} 1.823718{col 62}{space 1}    0.11{col 71}{space 3}0.910{col 79}{space 4} -3.36743{col 92}{space 3} 3.781413
{txt}{space 21}volgs13_los15pct {c |}{col 39}{res}{space 2}-3.845571{col 51}{space 2} 2.589536{col 62}{space 1}   -1.49{col 71}{space 3}0.138{col 79}{space 4}-8.920969{col 92}{space 3} 1.229826
{txt}{space 37} {c |}
{space 2}c.voluntary_gs13#c.volgs13_los15pct {c |}{col 39}{res}{space 2} 33.56285{col 51}{space 2} 15.41676{col 62}{space 1}    2.18{col 71}{space 3}0.029{col 79}{space 4} 3.346558{col 92}{space 3} 63.77914
{txt}{space 37} {c |}
{space 33}cio5 {c |}{col 39}{res}{space 2}-.1302147{col 51}{space 2} .0821074{col 62}{space 1}   -1.59{col 71}{space 3}0.113{col 79}{space 4}-.2911423{col 92}{space 3} .0307129
{txt}{space 16}activityno_permil_pct {c |}{col 39}{res}{space 2}  .310991{col 51}{space 2} .1977143{col 62}{space 1}    1.57{col 71}{space 3}0.116{col 79}{space 4}-.0765218{col 92}{space 3} .6985038
{txt}{space 27}cio_tenure {c |}{col 39}{res}{space 2}-.0059237{col 51}{space 2} .0238615{col 62}{space 1}   -0.25{col 71}{space 3}0.804{col 79}{space 4}-.0526913{col 92}{space 3}  .040844
{txt}{space 30}lnitemp {c |}{col 39}{res}{space 2} .1837772{col 51}{space 2} .2465907{col 62}{space 1}    0.75{col 71}{space 3}0.456{col 79}{space 4}-.2995317{col 92}{space 3} .6670861
{txt}{space 27}lnprojsize {c |}{col 39}{res}{space 2} .0830036{col 51}{space 2} .0317219{col 62}{space 1}    2.62{col 71}{space 3}0.009{col 79}{space 4} .0208299{col 92}{space 3} .1451774
{txt}{space 26}projects_no {c |}{col 39}{res}{space 2} .0007799{col 51}{space 2} .0018162{col 62}{space 1}    0.43{col 71}{space 3}0.668{col 79}{space 4}-.0027798{col 92}{space 3} .0043396
{txt}{space 30}accrate {c |}{col 39}{res}{space 2} .3551956{col 51}{space 2} .5477935{col 62}{space 1}    0.65{col 71}{space 3}0.517{col 79}{space 4}-.7184598{col 92}{space 3} 1.428851
{txt}{space 22}newprojects_pct {c |}{col 39}{res}{space 2}-.8180418{col 51}{space 2} .1657353{col 62}{space 1}   -4.94{col 71}{space 3}0.000{col 79}{space 4}-1.142877{col 92}{space 3}-.4932067
{txt}{space 35}y1 {c |}{col 39}{res}{space 2}-1.212652{col 51}{space 2} .1072233{col 62}{space 1}  -11.31{col 71}{space 3}0.000{col 79}{space 4}-1.422806{col 92}{space 3}-1.002498
{txt}{space 35}y2 {c |}{col 39}{res}{space 2}-1.178031{col 51}{space 2} .1301187{col 62}{space 1}   -9.05{col 71}{space 3}0.000{col 79}{space 4}-1.433059{col 92}{space 3}-.9230031
{txt}{space 35}y3 {c |}{col 39}{res}{space 2}-1.132377{col 51}{space 2} .1204601{col 62}{space 1}   -9.40{col 71}{space 3}0.000{col 79}{space 4}-1.368475{col 92}{space 3}-.8962797
{txt}{space 35}y4 {c |}{col 39}{res}{space 2}-.8694216{col 51}{space 2} .1180684{col 62}{space 1}   -7.36{col 71}{space 3}0.000{col 79}{space 4}-1.100831{col 92}{space 3}-.6380118
{txt}{space 35}y5 {c |}{col 39}{res}{space 2} -1.00039{col 51}{space 2} .1093205{col 62}{space 1}   -9.15{col 71}{space 3}0.000{col 79}{space 4}-1.214654{col 92}{space 3}-.7861257
{txt}{space 35}y6 {c |}{col 39}{res}{space 2}-.9141757{col 51}{space 2} .1029215{col 62}{space 1}   -8.88{col 71}{space 3}0.000{col 79}{space 4}-1.115898{col 92}{space 3}-.7124533
{txt}{space 35}y7 {c |}{col 39}{res}{space 2}        0{col 51}{txt}  (omitted)
{space 22}voluntary_gs13b {c |}{col 39}{res}{space 2} 2.512342{col 51}{space 2} 3.718323{col 62}{space 1}    0.68{col 71}{space 3}0.499{col 79}{space 4}-4.775437{col 92}{space 3} 9.800122
{txt}{space 20}volgs13_los15pctb {c |}{col 39}{res}{space 2} 9.744373{col 51}{space 2}   6.5832{col 62}{space 1}    1.48{col 71}{space 3}0.139{col 79}{space 4}-3.158462{col 92}{space 3} 22.64721
{txt}{space 37} {c |}
c.voluntary_gs13b#c.volgs13_los15pctb {c |}{col 39}{res}{space 2}-198.3789{col 51}{space 2} 73.97955{col 62}{space 1}   -2.68{col 71}{space 3}0.007{col 79}{space 4}-343.3762{col 92}{space 3}-53.38169
{txt}{space 37} {c |}
{space 15}activityno_permil_pctb {c |}{col 39}{res}{space 2}  1.03163{col 51}{space 2} .3949808{col 62}{space 1}    2.61{col 71}{space 3}0.009{col 79}{space 4} .2574817{col 92}{space 3} 1.805778
{txt}{space 29}lnitempb {c |}{col 39}{res}{space 2}-.1660587{col 51}{space 2} .2485683{col 62}{space 1}   -0.67{col 71}{space 3}0.504{col 79}{space 4}-.6532437{col 92}{space 3} .3211263
{txt}{space 26}lnprojsizeb {c |}{col 39}{res}{space 2}-.0033598{col 51}{space 2}  .065596{col 62}{space 1}   -0.05{col 71}{space 3}0.959{col 79}{space 4}-.1319256{col 92}{space 3} .1252059
{txt}{space 25}projects_nob {c |}{col 39}{res}{space 2}-.0049371{col 51}{space 2} .0033801{col 62}{space 1}   -1.46{col 71}{space 3}0.144{col 79}{space 4}-.0115621{col 92}{space 3} .0016878
{txt}{space 29}accrateb {c |}{col 39}{res}{space 2}-1.890788{col 51}{space 2} 1.447103{col 62}{space 1}   -1.31{col 71}{space 3}0.191{col 79}{space 4}-4.727057{col 92}{space 3} .9454821
{txt}{space 21}newprojects_pctb {c |}{col 39}{res}{space 2}-.9425394{col 51}{space 2} .3380202{col 62}{space 1}   -2.79{col 71}{space 3}0.005{col 79}{space 4}-1.605047{col 92}{space 3}-.2800321
{txt}{space 26}cio_tenureb {c |}{col 39}{res}{space 2}-.0392195{col 51}{space 2} .0555111{col 62}{space 1}   -0.71{col 71}{space 3}0.480{col 79}{space 4}-.1480192{col 92}{space 3} .0695801
{txt}{space 15}dmeratio_totalspending {c |}{col 39}{res}{space 2} .1608811{col 51}{space 2} .3759508{col 62}{space 1}    0.43{col 71}{space 3}0.669{col 79}{space 4} -.575969{col 92}{space 3} .8977312
{txt}{space 14}dmeratio_totalspendingb {c |}{col 39}{res}{space 2} .5475662{col 51}{space 2} .5305838{col 62}{space 1}    1.03{col 71}{space 3}0.302{col 79}{space 4}-.4923589{col 92}{space 3} 1.587491
{txt}{space 32}_cons {c |}{col 39}{res}{space 2} .6047265{col 51}{space 2} .3526116{col 62}{space 1}    1.71{col 71}{space 3}0.086{col 79}{space 4}-.0863796{col 92}{space 3} 1.295833
{txt}{hline 38}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est2{txt} stored)

{com}. 
. * Managerial Voluntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins , dydx(voluntary_gs13) at(volgs13_los15pct=($volgs13_los_p10 $volgs13_los_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 3:.05}}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}   Contrast{col 28} Delta-method{col 39}    Una{col 48}djusted{col 56}          Una{col 69}djusted
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 11}_at {c |}
{space 7}2 vs 1  {c |}{col 16}{res}{space 2} .4646149{col 28}{space 2} .1858773{col 39}{space 1}    2.50{col 48}{space 3}0.012{col 56}{space 4}  .100302{col 69}{space 3} .8289277
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 4E]
. margins , dydx(voluntary_gs13) at(volgs13_los15pct=(0(0.025)0.2))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:0}}
{lalign 7:2._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.025}}
{lalign 7:3._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.05}}
{lalign 7:4._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.075}}
{lalign 7:5._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.1}}
{lalign 7:6._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.125}}
{lalign 7:7._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.15}}
{lalign 7:8._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.175}}
{lalign 7:9._at: }{space 0}{lalign 16:volgs13_los15pct} = {res:{ralign 4:.2}}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}{col 28} Delta-method
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 11}_at {c |}
{space 12}1  {c |}{col 16}{res}{space 2} .0625608{col 28}{space 2} .5518281{col 39}{space 1}    0.11{col 48}{space 3}0.910{col 56}{space 4}-1.019002{col 69}{space 3} 1.144124
{txt}{space 12}2  {c |}{col 16}{res}{space 2} .3047608{col 28}{space 2} .5607572{col 39}{space 1}    0.54{col 48}{space 3}0.587{col 56}{space 4}-.7943031{col 69}{space 3} 1.403825
{txt}{space 12}3  {c |}{col 16}{res}{space 2} .5271757{col 28}{space 2} .5710194{col 39}{space 1}    0.92{col 48}{space 3}0.356{col 56}{space 4}-.5920017{col 69}{space 3} 1.646353
{txt}{space 12}4  {c |}{col 16}{res}{space 2} .7282449{col 28}{space 2}  .572408{col 39}{space 1}    1.27{col 48}{space 3}0.203{col 56}{space 4}-.3936542{col 69}{space 3} 1.850144
{txt}{space 12}5  {c |}{col 16}{res}{space 2} .9069467{col 28}{space 2} .5585932{col 39}{space 1}    1.62{col 48}{space 3}0.104{col 56}{space 4}-.1878758{col 69}{space 3} 2.001769
{txt}{space 12}6  {c |}{col 16}{res}{space 2} 1.062782{col 28}{space 2} .5267725{col 39}{space 1}    2.02{col 48}{space 3}0.044{col 56}{space 4} .0303271{col 69}{space 3} 2.095237
{txt}{space 12}7  {c |}{col 16}{res}{space 2} 1.195743{col 28}{space 2}  .477158{col 39}{space 1}    2.51{col 48}{space 3}0.012{col 56}{space 4}   .26053{col 69}{space 3} 2.130955
{txt}{space 12}8  {c |}{col 16}{res}{space 2} 1.306265{col 28}{space 2} .4129215{col 39}{space 1}    3.16{col 48}{space 3}0.002{col 56}{space 4} .4969535{col 69}{space 3} 2.115576
{txt}{space 12}9  {c |}{col 16}{res}{space 2} 1.395178{col 28}{space 2} .3412408{col 39}{space 1}    4.09{col 48}{space 3}0.000{col 56}{space 4}  .726358{col 69}{space 3} 2.063997
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// 
> xlabel(1 "1" 5 "5" 10 "10" ,format(%9.0f)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departing Managers' Service Length (>15 years, %)", margin(t=1)  size(11pt)) ///
> title("{c -(}bf: Figure 4E. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Managerial Voluntary Turnover Model [MODEL 18]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram volgs13_los15pct if volgs13_los15pct<=0.2 , yaxis(2) xaxis(1)  fcolor(gray%20) yscale(axis(2) alt) lc(gray%10) xlabel(0 "0" 0.1 "10" 0.2 "20" ,format(%9.0f)) ylabel( ,format(%9.1f) angle(0) ) ylabel(,format(%9.1f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:volgs13_los15pct}{p_end}
{res}{txt}
{com}. 
. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4E.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4E.04-29-2025.gph} saved

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 19, FIGURE 5E] H3: Turnover X Agency Centralization
. eststo: xtgee delay_pct c.voluntary_gs13##i.cio5 c.volgs13_los15pct  activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.voluntary_gs13b c.volgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}1.2968423
{txt}Iteration 2:  Tolerance = {res}.19898136
{txt}Iteration 3:  Tolerance = {res}.0723359
{txt}Iteration 4:  Tolerance = {res}.03116665
{txt}Iteration 5:  Tolerance = {res}.01134083
{txt}Iteration 6:  Tolerance = {res}.00456069
{txt}Iteration 7:  Tolerance = {res}.00198541
{txt}Iteration 8:  Tolerance = {res}.0008524
{txt}Iteration 9:  Tolerance = {res}.00036379
{txt}Iteration 10: Tolerance = {res}.00015491
{txt}Iteration 11: Tolerance = {res}.00006593
{txt}Iteration 12: Tolerance = {res}.00002807
{txt}Iteration 13: Tolerance = {res}.00001196
{txt}Iteration 14: Tolerance = {res}5.095e-06
{txt}Iteration 15: Tolerance = {res}2.173e-06
{txt}Iteration 16: Tolerance = {res}9.271e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:323.82}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}voluntary_gs13 {c |}{col 25}{res}{space 2}  .513452{col 37}{space 2} 1.845417{col 48}{space 1}    0.28{col 57}{space 3}0.781{col 65}{space 4}-3.103498{col 78}{space 3} 4.130402
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.0346934{col 37}{space 2}  .115574{col 48}{space 1}   -0.30{col 57}{space 3}0.764{col 65}{space 4}-.2612142{col 78}{space 3} .1918274
{txt}{space 23} {c |}
{space 2}cio5#c.voluntary_gs13 {c |}
{space 21}1  {c |}{col 25}{res}{space 2}-2.016273{col 37}{space 2} 3.104939{col 48}{space 1}   -0.65{col 57}{space 3}0.516{col 65}{space 4}-8.101843{col 78}{space 3} 4.069296
{txt}{space 23} {c |}
{space 7}volgs13_los15pct {c |}{col 25}{res}{space 2}-.1357666{col 37}{space 2} 1.029339{col 48}{space 1}   -0.13{col 57}{space 3}0.895{col 65}{space 4}-2.153234{col 78}{space 3} 1.881701
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2826788{col 37}{space 2} .1976414{col 48}{space 1}    1.43{col 57}{space 3}0.153{col 65}{space 4}-.1046912{col 78}{space 3} .6700488
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0146368{col 37}{space 2} .0242042{col 48}{space 1}   -0.60{col 57}{space 3}0.545{col 65}{space 4}-.0620762{col 78}{space 3} .0328026
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .1207873{col 37}{space 2} .2384433{col 48}{space 1}    0.51{col 57}{space 3}0.612{col 65}{space 4} -.346553{col 78}{space 3} .5881276
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0810891{col 37}{space 2} .0272063{col 48}{space 1}    2.98{col 57}{space 3}0.003{col 65}{space 4} .0277658{col 78}{space 3} .1344124
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0009869{col 37}{space 2} .0016683{col 48}{space 1}    0.59{col 57}{space 3}0.554{col 65}{space 4} -.002283{col 78}{space 3} .0042568
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .3328994{col 37}{space 2} .5433217{col 48}{space 1}    0.61{col 57}{space 3}0.540{col 65}{space 4}-.7319916{col 78}{space 3}  1.39779
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8071781{col 37}{space 2} .1652324{col 48}{space 1}   -4.89{col 57}{space 3}0.000{col 65}{space 4}-1.131028{col 78}{space 3}-.4833286
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.195244{col 37}{space 2} .1044526{col 48}{space 1}  -11.44{col 57}{space 3}0.000{col 65}{space 4}-1.399967{col 78}{space 3}-.9905203
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.177495{col 37}{space 2} .1266778{col 48}{space 1}   -9.30{col 57}{space 3}0.000{col 65}{space 4}-1.425779{col 78}{space 3}-.9292114
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.120193{col 37}{space 2} .1181979{col 48}{space 1}   -9.48{col 57}{space 3}0.000{col 65}{space 4}-1.351857{col 78}{space 3}-.8885298
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.8778718{col 37}{space 2} .1222094{col 48}{space 1}   -7.18{col 57}{space 3}0.000{col 65}{space 4}-1.117398{col 78}{space 3}-.6383459
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9951909{col 37}{space 2}  .109547{col 48}{space 1}   -9.08{col 57}{space 3}0.000{col 65}{space 4}-1.209899{col 78}{space 3}-.7804827
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8967892{col 37}{space 2} .1023627{col 48}{space 1}   -8.76{col 57}{space 3}0.000{col 65}{space 4}-1.097416{col 78}{space 3} -.696162
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 8}voluntary_gs13b {c |}{col 25}{res}{space 2} .0068794{col 37}{space 2}  3.25477{col 48}{space 1}    0.00{col 57}{space 3}0.998{col 65}{space 4}-6.372353{col 78}{space 3} 6.386112
{txt}{space 6}volgs13_los15pctb {c |}{col 25}{res}{space 2}-6.576766{col 37}{space 2} 2.311254{col 48}{space 1}   -2.85{col 57}{space 3}0.004{col 65}{space 4}-11.10674{col 78}{space 3}-2.046792
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .9929336{col 37}{space 2} .4088709{col 48}{space 1}    2.43{col 57}{space 3}0.015{col 65}{space 4} .1915614{col 78}{space 3} 1.794306
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2} -.115128{col 37}{space 2} .2437837{col 48}{space 1}   -0.47{col 57}{space 3}0.637{col 65}{space 4}-.5929352{col 78}{space 3} .3626793
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2} .0021299{col 37}{space 2} .0635885{col 48}{space 1}    0.03{col 57}{space 3}0.973{col 65}{space 4}-.1225012{col 78}{space 3}  .126761
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0055509{col 37}{space 2} .0032705{col 48}{space 1}   -1.70{col 57}{space 3}0.090{col 65}{space 4}-.0119609{col 78}{space 3} .0008591
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-.8785867{col 37}{space 2} 1.239383{col 48}{space 1}   -0.71{col 57}{space 3}0.478{col 65}{space 4}-3.307733{col 78}{space 3}  1.55056
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-1.016766{col 37}{space 2} .3389303{col 48}{space 1}   -3.00{col 57}{space 3}0.003{col 65}{space 4}-1.681057{col 78}{space 3}-.3524749
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0313605{col 37}{space 2} .0563319{col 48}{space 1}   -0.56{col 57}{space 3}0.578{col 65}{space 4}-.1417689{col 78}{space 3} .0790479
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .3085606{col 37}{space 2} .4295809{col 48}{space 1}    0.72{col 57}{space 3}0.473{col 65}{space 4}-.5334025{col 78}{space 3} 1.150524
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .4141892{col 37}{space 2} .5542267{col 48}{space 1}    0.75{col 57}{space 3}0.455{col 65}{space 4}-.6720752{col 78}{space 3} 1.500454
{txt}{space 18}_cons {c |}{col 25}{res}{space 2}  .758214{col 37}{space 2} .3651297{col 48}{space 1}    2.08{col 57}{space 3}0.038{col 65}{space 4} .0425729{col 78}{space 3} 1.473855
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est3{txt} stored)

{com}. 
. * Managerial Voluntary Turnover Rate: Marginal Effect Differentials Based on Agency Centralization
. margins , dydx(voluntary_gs13) over(cio5) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}   Contrast{col 28} Delta-method{col 39}    Una{col 48}djusted{col 56}          Una{col 69}djusted
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 10}cio5 {c |}
{space 7}1 vs 0  {c |}{col 16}{res}{space 2}-.5877579{col 28}{space 2} .8954925{col 39}{space 1}   -0.66{col 48}{space 3}0.512{col 56}{space 4}-2.342891{col 69}{space 3} 1.167375
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 5E]
. margins , dydx(voluntary_gs13) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}{col 28} Delta-method
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 10}cio5 {c |}
{space 12}0  {c |}{col 16}{res}{space 2} .1550489{col 28}{space 2} .5578455{col 39}{space 1}    0.28{col 48}{space 3}0.781{col 56}{space 4}-.9383081{col 69}{space 3} 1.248406
{txt}{space 12}1  {c |}{col 16}{res}{space 2}-.4327089{col 28}{space 2} 1.015999{col 39}{space 1}   -0.43{col 48}{space 3}0.670{col 56}{space 4}-2.424031{col 69}{space 3} 1.558613
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(black)) ciopt(color(black%60))  yline(0, lcolor(red) lpattern(shortdash)) ///
> xlabel (-1 " " 0 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5E. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Managerial Voluntary Turnover Model [MODEL 19]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5E.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5E.04-29-2025.gph} saved

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. * [MODEL 20, FIGURE 6E] H4 : Turnover X Task Standardization
. eststo: xtgee delay_pct c.voluntary_gs13##c.activityno_permil_pct i.cio5 c.volgs13_los15pct   cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.voluntary_gs13b c.volgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}.33156152
{txt}Iteration 2:  Tolerance = {res}.10870878
{txt}Iteration 3:  Tolerance = {res}.03946801
{txt}Iteration 4:  Tolerance = {res}.01623292
{txt}Iteration 5:  Tolerance = {res}.00730158
{txt}Iteration 6:  Tolerance = {res}.00343236
{txt}Iteration 7:  Tolerance = {res}.00154801
{txt}Iteration 8:  Tolerance = {res}.00068338
{txt}Iteration 9:  Tolerance = {res}.00029836
{txt}Iteration 10: Tolerance = {res}.00012955
{txt}Iteration 11: Tolerance = {res}.00005612
{txt}Iteration 12: Tolerance = {res}.00002429
{txt}Iteration 13: Tolerance = {res}.00001052
{txt}Iteration 14: Tolerance = {res}4.556e-06
{txt}Iteration 15: Tolerance = {res}1.975e-06
{txt}Iteration 16: Tolerance = {res}8.567e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:333.90}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 106:(Std. err. adjusted for clustering on {res:a_id})}
{hline 41}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 42}{c |}{col 54}  Semirobust
{col 1}                               delay_pct{col 42}{c |} Coefficient{col 54}  std. err.{col 66}      z{col 74}   P>|z|{col 82}     [95% con{col 95}f. interval]
{hline 41}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}voluntary_gs13 {c |}{col 42}{res}{space 2} 2.604728{col 54}{space 2} 2.754511{col 65}{space 1}    0.95{col 74}{space 3}0.344{col 82}{space 4}-2.794014{col 95}{space 3} 8.003469
{txt}{space 19}activityno_permil_pct {c |}{col 42}{res}{space 2} .4053512{col 54}{space 2} .2312613{col 65}{space 1}    1.75{col 74}{space 3}0.080{col 82}{space 4}-.0479127{col 95}{space 3} .8586151
{txt}{space 40} {c |}
c.voluntary_gs13#c.activityno_permil_pct {c |}{col 42}{res}{space 2}-5.291075{col 54}{space 2} 3.847896{col 65}{space 1}   -1.38{col 74}{space 3}0.169{col 82}{space 4}-12.83281{col 95}{space 3} 2.250662
{txt}{space 40} {c |}
{space 34}1.cio5 {c |}{col 42}{res}{space 2}-.0934568{col 54}{space 2} .0825662{col 65}{space 1}   -1.13{col 74}{space 3}0.258{col 82}{space 4}-.2552836{col 95}{space 3}   .06837
{txt}{space 24}volgs13_los15pct {c |}{col 42}{res}{space 2}-.6658979{col 54}{space 2} 1.305077{col 65}{space 1}   -0.51{col 74}{space 3}0.610{col 82}{space 4}-3.223801{col 95}{space 3} 1.892005
{txt}{space 30}cio_tenure {c |}{col 42}{res}{space 2}-.0094907{col 54}{space 2}  .024241{col 65}{space 1}   -0.39{col 74}{space 3}0.695{col 82}{space 4}-.0570023{col 95}{space 3} .0380209
{txt}{space 33}lnitemp {c |}{col 42}{res}{space 2} .0650215{col 54}{space 2} .2206547{col 65}{space 1}    0.29{col 74}{space 3}0.768{col 82}{space 4}-.3674537{col 95}{space 3} .4974967
{txt}{space 30}lnprojsize {c |}{col 42}{res}{space 2} .0798253{col 54}{space 2} .0301354{col 65}{space 1}    2.65{col 74}{space 3}0.008{col 82}{space 4} .0207611{col 95}{space 3} .1388895
{txt}{space 29}projects_no {c |}{col 42}{res}{space 2} .0012508{col 54}{space 2} .0017564{col 65}{space 1}    0.71{col 74}{space 3}0.476{col 82}{space 4}-.0021917{col 95}{space 3} .0046933
{txt}{space 33}accrate {c |}{col 42}{res}{space 2} .4070993{col 54}{space 2} .5441435{col 65}{space 1}    0.75{col 74}{space 3}0.454{col 82}{space 4}-.6594023{col 95}{space 3} 1.473601
{txt}{space 25}newprojects_pct {c |}{col 42}{res}{space 2} -.828192{col 54}{space 2} .1654183{col 65}{space 1}   -5.01{col 74}{space 3}0.000{col 82}{space 4}-1.152406{col 95}{space 3} -.503978
{txt}{space 38}y1 {c |}{col 42}{res}{space 2}-1.212948{col 54}{space 2} .1042544{col 65}{space 1}  -11.63{col 74}{space 3}0.000{col 82}{space 4}-1.417283{col 95}{space 3}-1.008613
{txt}{space 38}y2 {c |}{col 42}{res}{space 2}-1.195544{col 54}{space 2} .1262725{col 65}{space 1}   -9.47{col 74}{space 3}0.000{col 82}{space 4}-1.443034{col 95}{space 3}-.9480549
{txt}{space 38}y3 {c |}{col 42}{res}{space 2}-1.132466{col 54}{space 2} .1178563{col 65}{space 1}   -9.61{col 74}{space 3}0.000{col 82}{space 4} -1.36346{col 95}{space 3}-.9014718
{txt}{space 38}y4 {c |}{col 42}{res}{space 2}-.8769647{col 54}{space 2} .1199821{col 65}{space 1}   -7.31{col 74}{space 3}0.000{col 82}{space 4}-1.112125{col 95}{space 3}-.6418041
{txt}{space 38}y5 {c |}{col 42}{res}{space 2} -1.01059{col 54}{space 2} .1097165{col 65}{space 1}   -9.21{col 74}{space 3}0.000{col 82}{space 4}-1.225631{col 95}{space 3}-.7955498
{txt}{space 38}y6 {c |}{col 42}{res}{space 2}-.9074722{col 54}{space 2} .1022033{col 65}{space 1}   -8.88{col 74}{space 3}0.000{col 82}{space 4}-1.107787{col 95}{space 3}-.7071574
{txt}{space 38}y7 {c |}{col 42}{res}{space 2}        0{col 54}{txt}  (omitted)
{space 25}voluntary_gs13b {c |}{col 42}{res}{space 2}-.2888696{col 54}{space 2} 3.306294{col 65}{space 1}   -0.09{col 74}{space 3}0.930{col 82}{space 4}-6.769087{col 95}{space 3} 6.191348
{txt}{space 23}volgs13_los15pctb {c |}{col 42}{res}{space 2}-6.428768{col 54}{space 2} 2.235114{col 65}{space 1}   -2.88{col 74}{space 3}0.004{col 82}{space 4}-10.80951{col 95}{space 3}-2.048025
{txt}{space 18}activityno_permil_pctb {c |}{col 42}{res}{space 2} 1.024509{col 54}{space 2} .3993826{col 65}{space 1}    2.57{col 74}{space 3}0.010{col 82}{space 4} .2417339{col 95}{space 3} 1.807285
{txt}{space 32}lnitempb {c |}{col 42}{res}{space 2}-.0583129{col 54}{space 2} .2272351{col 65}{space 1}   -0.26{col 74}{space 3}0.797{col 82}{space 4}-.5036855{col 95}{space 3} .3870597
{txt}{space 29}lnprojsizeb {c |}{col 42}{res}{space 2} .0054885{col 54}{space 2} .0632447{col 65}{space 1}    0.09{col 74}{space 3}0.931{col 82}{space 4}-.1184689{col 95}{space 3} .1294459
{txt}{space 28}projects_nob {c |}{col 42}{res}{space 2}-.0056335{col 54}{space 2} .0032923{col 65}{space 1}   -1.71{col 74}{space 3}0.087{col 82}{space 4}-.0120863{col 95}{space 3} .0008194
{txt}{space 32}accrateb {c |}{col 42}{res}{space 2}-.9983269{col 54}{space 2}  1.23884{col 65}{space 1}   -0.81{col 74}{space 3}0.420{col 82}{space 4}-3.426408{col 95}{space 3} 1.429754
{txt}{space 24}newprojects_pctb {c |}{col 42}{res}{space 2}-.9914435{col 54}{space 2} .3340103{col 65}{space 1}   -2.97{col 74}{space 3}0.003{col 82}{space 4}-1.646092{col 95}{space 3}-.3367954
{txt}{space 29}cio_tenureb {c |}{col 42}{res}{space 2}-.0358079{col 54}{space 2} .0556781{col 65}{space 1}   -0.64{col 74}{space 3}0.520{col 82}{space 4} -.144935{col 95}{space 3} .0733192
{txt}{space 18}dmeratio_totalspending {c |}{col 42}{res}{space 2} .3042974{col 54}{space 2} .4149828{col 65}{space 1}    0.73{col 74}{space 3}0.463{col 82}{space 4} -.509054{col 95}{space 3} 1.117649
{txt}{space 17}dmeratio_totalspendingb {c |}{col 42}{res}{space 2} .3776896{col 54}{space 2} .5358598{col 65}{space 1}    0.70{col 74}{space 3}0.481{col 82}{space 4}-.6725764{col 95}{space 3} 1.427956
{txt}{space 35}_cons {c |}{col 42}{res}{space 2} .7166761{col 54}{space 2} .3584485{col 65}{space 1}    2.00{col 74}{space 3}0.046{col 82}{space 4}   .01413{col 95}{space 3} 1.419222
{txt}{hline 41}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est4{txt} stored)

{com}. 
. * Managerial Voluntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Task Standardization
. margins , dydx(voluntary_gs13) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}   Contrast{col 28} Delta-method{col 39}    Una{col 48}djusted{col 56}          Una{col 69}djusted
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 11}_at {c |}
{space 7}2 vs 1  {c |}{col 16}{res}{space 2}-1.112012{col 28}{space 2} .8021897{col 39}{space 1}   -1.39{col 48}{space 3}0.166{col 56}{space 4}-2.684275{col 69}{space 3} .4602508
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. 
. * [FIGURE 6E]
. margins , dydx(voluntary_gs13) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:voluntary_gs13}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 15}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 16}{c |}{col 28} Delta-method
{col 16}{c |}      dy/dx{col 28}   std. err.{col 40}      z{col 48}   P>|z|{col 56}     [95% con{col 69}f. interval]
{hline 15}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}voluntary_gs13 {txt}{c |}
{space 11}_at {c |}
{space 12}1  {c |}{col 16}{res}{space 2} .7506708{col 28}{space 2} .7941822{col 39}{space 1}    0.95{col 48}{space 3}0.345{col 56}{space 4}-.8058977{col 69}{space 3} 2.307239
{txt}{space 12}2  {c |}{col 16}{res}{space 2} .6065113{col 28}{space 2} .7239378{col 39}{space 1}    0.84{col 48}{space 3}0.402{col 56}{space 4}-.8123808{col 69}{space 3} 2.025403
{txt}{space 12}3  {c |}{col 16}{res}{space 2} .4579157{col 28}{space 2} .6604035{col 39}{space 1}    0.69{col 48}{space 3}0.488{col 56}{space 4}-.8364513{col 69}{space 3} 1.752283
{txt}{space 12}4  {c |}{col 16}{res}{space 2} .3050707{col 28}{space 2} .6074866{col 39}{space 1}    0.50{col 48}{space 3}0.616{col 56}{space 4}-.8855812{col 69}{space 3} 1.495723
{txt}{space 12}5  {c |}{col 16}{res}{space 2} .1481821{col 28}{space 2} .5700488{col 39}{space 1}    0.26{col 48}{space 3}0.795{col 56}{space 4}-.9690931{col 69}{space 3} 1.265457
{txt}{space 12}6  {c |}{col 16}{res}{space 2}-.0125259{col 28}{space 2} .5531261{col 39}{space 1}   -0.02{col 48}{space 3}0.982{col 56}{space 4}-1.096633{col 69}{space 3} 1.071581
{txt}{space 12}7  {c |}{col 16}{res}{space 2}-.1768101{col 28}{space 2} .5603145{col 39}{space 1}   -0.32{col 48}{space 3}0.752{col 56}{space 4}-1.275006{col 69}{space 3} .9213861
{txt}{space 12}8  {c |}{col 16}{res}{space 2}-.3444095{col 28}{space 2}  .592206{col 39}{space 1}   -0.58{col 48}{space 3}0.561{col 56}{space 4}-1.505112{col 69}{space 3}  .816293
{txt}{space 12}9  {c |}{col 16}{res}{space 2} -.515046{col 28}{space 2} .6463018{col 39}{space 1}   -0.80{col 48}{space 3}0.426{col 56}{space 4}-1.781774{col 69}{space 3} .7516823
{txt}{space 11}10  {c |}{col 16}{res}{space 2}-.6884242{col 28}{space 2} .7184224{col 39}{space 1}   -0.96{col 48}{space 3}0.338{col 56}{space 4}-2.096506{col 69}{space 3} .7196578
{txt}{space 11}11  {c |}{col 16}{res}{space 2}-.8642337{col 28}{space 2} .8042443{col 39}{space 1}   -1.07{col 48}{space 3}0.283{col 56}{space 4}-2.440524{col 69}{space 3} .7120562
{txt}{hline 15}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", size(11pt) margin(t=1)) ///
> title("{c -(}bf: Figure 6E. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Managerial Voluntary Turnover Model [MODEL 20]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f))  ylabel(0 "0" 10 "10" 20 "20",format(%9.1f) angle(0) axis(2)) ytitle(,  axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6E.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6E.04-29-2025.gph} saved

{com}. 
. 
. 
. ************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. 
. 
. **# [MODELS 21 - 24, FIGURES 3F - 6F] MANAGERIAL INVOLUNTARY TURNOVER MODELS
. 
. 
. quietly: xtgee delay_pct c.turnover_gs13  gs13above15_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 turnover_gs13b gs13above15_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb, fam(bin) link(probit) corr(uns) vce(robust)
{txt}
{com}. 
. * [MODEL 21, FIGURE 3F] H1: Managerial Involuntary Turnover Baseline Effects
. eststo: xtgee delay_pct c.involuntary_gs13 c.involgs13_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.involuntary_gs13b  c.involgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}7.3729754
{txt}Iteration 2:  Tolerance = {res}.36031379
{txt}Iteration 3:  Tolerance = {res}.32160414
{txt}Iteration 4:  Tolerance = {res}.30372585
{txt}Iteration 5:  Tolerance = {res}.26292866
{txt}Iteration 6:  Tolerance = {res}.20793709
{txt}Iteration 7:  Tolerance = {res}.15123445
{txt}Iteration 8:  Tolerance = {res}.10215504
{txt}Iteration 9:  Tolerance = {res}.065103
{txt}Iteration 10: Tolerance = {res}.03984885
{txt}Iteration 11: Tolerance = {res}.0237757
{txt}Iteration 12: Tolerance = {res}.01397117
{txt}Iteration 13: Tolerance = {res}.00813889
{txt}Iteration 14: Tolerance = {res}.00471893
{txt}Iteration 15: Tolerance = {res}.00272932
{txt}Iteration 16: Tolerance = {res}.00157669
{txt}Iteration 17: Tolerance = {res}.00091037
{txt}Iteration 18: Tolerance = {res}.00052556
{txt}Iteration 19: Tolerance = {res}.00030342
{txt}Iteration 20: Tolerance = {res}.00017518
{txt}Iteration 21: Tolerance = {res}.00010116
{txt}Iteration 22: Tolerance = {res}.00005842
{txt}Iteration 23: Tolerance = {res}.00003374
{txt}Iteration 24: Tolerance = {res}.00001949
{txt}Iteration 25: Tolerance = {res}.00001126
{txt}Iteration 26: Tolerance = {res}6.505e-06
{txt}Iteration 27: Tolerance = {res}3.758e-06
{txt}Iteration 28: Tolerance = {res}2.171e-06
{txt}Iteration 29: Tolerance = {res}1.255e-06
{txt}Iteration 30: Tolerance = {res}7.249e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:296.96}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 7}involuntary_gs13 {c |}{col 25}{res}{space 2}-.9004654{col 37}{space 2} 9.075284{col 48}{space 1}   -0.10{col 57}{space 3}0.921{col 65}{space 4} -18.6877{col 78}{space 3} 16.88676
{txt}{space 5}involgs13_los15pct {c |}{col 25}{res}{space 2}-1.004325{col 37}{space 2} 13.68034{col 48}{space 1}   -0.07{col 57}{space 3}0.941{col 65}{space 4}-27.81731{col 78}{space 3} 25.80866
{txt}{space 19}cio5 {c |}{col 25}{res}{space 2}-.1611636{col 37}{space 2} .0845595{col 48}{space 1}   -1.91{col 57}{space 3}0.057{col 65}{space 4}-.3268971{col 78}{space 3}   .00457
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2548252{col 37}{space 2} .1974594{col 48}{space 1}    1.29{col 57}{space 3}0.197{col 65}{space 4}-.1321882{col 78}{space 3} .6418386
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2}-.0106166{col 37}{space 2} .0241412{col 48}{space 1}   -0.44{col 57}{space 3}0.660{col 65}{space 4}-.0579324{col 78}{space 3} .0366992
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .2445108{col 37}{space 2} .2666967{col 48}{space 1}    0.92{col 57}{space 3}0.359{col 65}{space 4}-.2782052{col 78}{space 3} .7672268
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0676632{col 37}{space 2} .0334384{col 48}{space 1}    2.02{col 57}{space 3}0.043{col 65}{space 4} .0021251{col 78}{space 3} .1332014
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2}  .001409{col 37}{space 2} .0019147{col 48}{space 1}    0.74{col 57}{space 3}0.462{col 65}{space 4}-.0023438{col 78}{space 3} .0051618
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .5791958{col 37}{space 2} .5703038{col 48}{space 1}    1.02{col 57}{space 3}0.310{col 65}{space 4}-.5385791{col 78}{space 3} 1.696971
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8449463{col 37}{space 2}   .17523{col 48}{space 1}   -4.82{col 57}{space 3}0.000{col 65}{space 4}-1.188391{col 78}{space 3}-.5015018
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.166929{col 37}{space 2} .1070751{col 48}{space 1}  -10.90{col 57}{space 3}0.000{col 65}{space 4}-1.376793{col 78}{space 3}-.9570659
{txt}{space 21}y2 {c |}{col 25}{res}{space 2} -1.15764{col 37}{space 2} .1273339{col 48}{space 1}   -9.09{col 57}{space 3}0.000{col 65}{space 4} -1.40721{col 78}{space 3}  -.90807
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.093841{col 37}{space 2} .1186001{col 48}{space 1}   -9.22{col 57}{space 3}0.000{col 65}{space 4}-1.326293{col 78}{space 3}-.8613893
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.8206566{col 37}{space 2} .1204177{col 48}{space 1}   -6.82{col 57}{space 3}0.000{col 65}{space 4}-1.056671{col 78}{space 3}-.5846423
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9573085{col 37}{space 2} .1112862{col 48}{space 1}   -8.60{col 57}{space 3}0.000{col 65}{space 4}-1.175425{col 78}{space 3}-.7391916
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8964065{col 37}{space 2} .1042628{col 48}{space 1}   -8.60{col 57}{space 3}0.000{col 65}{space 4}-1.100758{col 78}{space 3}-.6920552
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 6}involuntary_gs13b {c |}{col 25}{res}{space 2} 44.85875{col 37}{space 2} 30.98476{col 48}{space 1}    1.45{col 57}{space 3}0.148{col 65}{space 4}-15.87026{col 78}{space 3} 105.5878
{txt}{space 4}involgs13_los15pctb {c |}{col 25}{res}{space 2} 32.53701{col 37}{space 2} 53.89426{col 48}{space 1}    0.60{col 57}{space 3}0.546{col 65}{space 4}-73.09381{col 78}{space 3} 138.1678
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8810768{col 37}{space 2} .3872812{col 48}{space 1}    2.28{col 57}{space 3}0.023{col 65}{space 4} .1220195{col 78}{space 3} 1.640134
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2}-.1947464{col 37}{space 2} .2695737{col 48}{space 1}   -0.72{col 57}{space 3}0.470{col 65}{space 4}-.7231012{col 78}{space 3} .3336084
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2}-.0201731{col 37}{space 2} .0699569{col 48}{space 1}   -0.29{col 57}{space 3}0.773{col 65}{space 4}-.1572861{col 78}{space 3} .1169399
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0036874{col 37}{space 2}  .002918{col 48}{space 1}   -1.26{col 57}{space 3}0.206{col 65}{space 4}-.0094066{col 78}{space 3} .0020318
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-1.535599{col 37}{space 2} 1.277114{col 48}{space 1}   -1.20{col 57}{space 3}0.229{col 65}{space 4}-4.038697{col 78}{space 3} .9674986
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-.7834573{col 37}{space 2}  .393858{col 48}{space 1}   -1.99{col 57}{space 3}0.047{col 65}{space 4}-1.555405{col 78}{space 3}-.0115098
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0108065{col 37}{space 2} .0591023{col 48}{space 1}   -0.18{col 57}{space 3}0.855{col 65}{space 4}-.1266449{col 78}{space 3}  .105032
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2} .0879215{col 37}{space 2} .4623784{col 48}{space 1}    0.19{col 57}{space 3}0.849{col 65}{space 4}-.8183236{col 78}{space 3} .9941666
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2}  .286597{col 37}{space 2} .5257073{col 48}{space 1}    0.55{col 57}{space 3}0.586{col 65}{space 4}-.7437704{col 78}{space 3} 1.316964
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .4219345{col 37}{space 2} .4425739{col 48}{space 1}    0.95{col 57}{space 3}0.340{col 65}{space 4}-.4454944{col 78}{space 3} 1.289363
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est5{txt} stored)

{com}. *
. *Managerial Involuntary Turnover Rate: Unconditional Marginal Effect 
. margins, dydx(involuntary_gs13)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2colreset}{...}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}{col 30} Delta-method
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
involuntary_gs13 {c |}{col 18}{res}{space 2}-.2723819{col 30}{space 2} 2.744757{col 41}{space 1}   -0.10{col 50}{space 3}0.921{col 58}{space 4}-5.652006{col 71}{space 3} 5.107242
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * Predicted Change in Delays Based on Interdecile Increase in Managerial Involuntary Turnover Rate
. margins, at(involuntary_gs13=($involuntary_gs13_p10 $involuntary_gs13_p90)) pwcompare(effects)
{txt}{p 0 6 2}note: ignoring pwcompare options because there are no margins for making pairwise comparisons.{p_end}
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 4:At: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 1:0}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 7}_cons {c |}{col 14}{res}{space 2} .2923225{col 26}{space 2} .0150482{col 37}{space 1}   19.43{col 46}{space 3}0.000{col 54}{space 4} .2628287{col 67}{space 3} .3218164
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 3F]
. margins, at(involuntary_gs13=(0(0.005)0.04)) 
{res}
{txt}{col 1}Predictive margins{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:0}}
{lalign 7:2._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.005}}
{lalign 7:3._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.01}}
{lalign 7:4._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.015}}
{lalign 7:5._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.02}}
{lalign 7:6._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.025}}
{lalign 7:7._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.03}}
{lalign 7:8._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.035}}
{lalign 7:9._at: }{space 0}{lalign 16:involuntary_gs13} = {res:{ralign 4:.04}}

{res}{txt}{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26} Delta-method
{col 14}{c |}     Margin{col 26}   std. err.{col 38}      z{col 46}   P>|z|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}_at {c |}
{space 10}1  {c |}{col 14}{res}{space 2} .2923225{col 26}{space 2} .0150482{col 37}{space 1}   19.43{col 46}{space 3}0.000{col 54}{space 4} .2628287{col 67}{space 3} .3218164
{txt}{space 10}2  {c |}{col 14}{res}{space 2}  .290962{col 26}{space 2} .0194598{col 37}{space 1}   14.95{col 46}{space 3}0.000{col 54}{space 4} .2528214{col 67}{space 3} .3291026
{txt}{space 10}3  {c |}{col 14}{res}{space 2} .2896046{col 26}{space 2} .0300466{col 37}{space 1}    9.64{col 46}{space 3}0.000{col 54}{space 4} .2307144{col 67}{space 3} .3484948
{txt}{space 10}4  {c |}{col 14}{res}{space 2} .2882503{col 26}{space 2} .0423507{col 37}{space 1}    6.81{col 46}{space 3}0.000{col 54}{space 4} .2052444{col 67}{space 3} .3712562
{txt}{space 10}5  {c |}{col 14}{res}{space 2} .2868992{col 26}{space 2}  .055189{col 37}{space 1}    5.20{col 46}{space 3}0.000{col 54}{space 4} .1787307{col 67}{space 3} .3950676
{txt}{space 10}6  {c |}{col 14}{res}{space 2} .2855512{col 26}{space 2} .0682224{col 37}{space 1}    4.19{col 46}{space 3}0.000{col 54}{space 4} .1518378{col 67}{space 3} .4192646
{txt}{space 10}7  {c |}{col 14}{res}{space 2} .2842064{col 26}{space 2} .0813248{col 37}{space 1}    3.49{col 46}{space 3}0.000{col 54}{space 4} .1248128{col 67}{space 3} .4436001
{txt}{space 10}8  {c |}{col 14}{res}{space 2} .2828649{col 26}{space 2} .0944396{col 37}{space 1}    3.00{col 46}{space 3}0.003{col 54}{space 4} .0977666{col 67}{space 3} .4679631
{txt}{space 10}9  {c |}{col 14}{res}{space 2} .2815265{col 26}{space 2} .1075378{col 37}{space 1}    2.62{col 46}{space 3}0.009{col 54}{space 4} .0707563{col 67}{space 3} .4922966
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) recastci(rarea) ///
> xlabel(0 "0" 0.02 "2" 0.04 "4", format(%9.0f) ) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0))  legend(off) ///
> ytitle("Predicted Project Delays (%)", size(11pt) margin(r=1)) xtitle("Managerial Involuntary Turnover (%)", margin(t=1)  size(11pt) ) ///
> title("{c -(}bf: Figure 3F. Unconditional Turnover Effects on Project Delays (H1){c )-}" "{c -(}bf:Managerial Involuntary Turnover Model [MODEL 21]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram involuntary if involuntary<=0.04, fc(gray%20) lc(gray%10) yaxis(2) yscale(r(0 20) axis(2) alt) xlabel(0 "0" 0.02 "2" 0.04 "4",format(%9.0f)) ylabel(0.1 "10" 0.2 "20" 0.3 "30" 0.4 "40" 0.5 "50",format(%9.0f) angle(0))  ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:involuntary_gs13}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3F.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 3F.04-29-2025.gph} saved

{com}. *
. *
. * [MODEL 22, FIGURE 4F] H2: Turnover X Service Length
. eststo: xtgee delay_pct c.involuntary_gs13##c.involgs13_los15pct cio5 activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.involuntary_gs13b  c.involgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}5.0321935
{txt}Iteration 2:  Tolerance = {res}.37594633
{txt}Iteration 3:  Tolerance = {res}.3529383
{txt}Iteration 4:  Tolerance = {res}.34319686
{txt}Iteration 5:  Tolerance = {res}.30847675
{txt}Iteration 6:  Tolerance = {res}.25459655
{txt}Iteration 7:  Tolerance = {res}.19337255
{txt}Iteration 8:  Tolerance = {res}.13571009
{txt}Iteration 9:  Tolerance = {res}.0890527
{txt}Iteration 10: Tolerance = {res}.05561104
{txt}Iteration 11: Tolerance = {res}.03361278
{txt}Iteration 12: Tolerance = {res}.01991708
{txt}Iteration 13: Tolerance = {res}.01166821
{txt}Iteration 14: Tolerance = {res}.00679356
{txt}Iteration 15: Tolerance = {res}.00394288
{txt}Iteration 16: Tolerance = {res}.00228496
{txt}Iteration 17: Tolerance = {res}.00132338
{txt}Iteration 18: Tolerance = {res}.00076634
{txt}Iteration 19: Tolerance = {res}.00044381
{txt}Iteration 20: Tolerance = {res}.00025706
{txt}Iteration 21: Tolerance = {res}.00014891
{txt}Iteration 22: Tolerance = {res}.00008628
{txt}Iteration 23: Tolerance = {res}.00005
{txt}Iteration 24: Tolerance = {res}.00002898
{txt}Iteration 25: Tolerance = {res}.0000168
{txt}Iteration 26: Tolerance = {res}9.736e-06
{txt}Iteration 27: Tolerance = {res}5.644e-06
{txt}Iteration 28: Tolerance = {res}3.272e-06
{txt}Iteration 29: Tolerance = {res}1.897e-06
{txt}Iteration 30: Tolerance = {res}1.100e-06
{txt}Iteration 31: Tolerance = {res}6.376e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:400.56}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 105:(Std. err. adjusted for clustering on {res:a_id})}
{hline 40}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 41}{c |}{col 53}  Semirobust
{col 1}                              delay_pct{col 41}{c |} Coefficient{col 53}  std. err.{col 65}      z{col 73}   P>|z|{col 81}     [95% con{col 94}f. interval]
{hline 40}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 23}involuntary_gs13 {c |}{col 41}{res}{space 2}-.1316998{col 53}{space 2} 8.815092{col 64}{space 1}   -0.01{col 73}{space 3}0.988{col 81}{space 4}-17.40896{col 94}{space 3} 17.14556
{txt}{space 21}involgs13_los15pct {c |}{col 41}{res}{space 2} 23.93112{col 53}{space 2} 23.66148{col 64}{space 1}    1.01{col 73}{space 3}0.312{col 81}{space 4}-22.44452{col 94}{space 3} 70.30676
{txt}{space 39} {c |}
c.involuntary_gs13#c.involgs13_los15pct {c |}{col 41}{res}{space 2} -1722.31{col 53}{space 2} 1186.566{col 64}{space 1}   -1.45{col 73}{space 3}0.147{col 81}{space 4}-4047.936{col 94}{space 3} 603.3167
{txt}{space 39} {c |}
{space 35}cio5 {c |}{col 41}{res}{space 2}-.1657655{col 53}{space 2}  .084013{col 64}{space 1}   -1.97{col 73}{space 3}0.048{col 81}{space 4} -.330428{col 94}{space 3} -.001103
{txt}{space 18}activityno_permil_pct {c |}{col 41}{res}{space 2} .2614564{col 53}{space 2} .1990921{col 64}{space 1}    1.31{col 73}{space 3}0.189{col 81}{space 4}-.1287569{col 94}{space 3} .6516697
{txt}{space 29}cio_tenure {c |}{col 41}{res}{space 2}-.0111356{col 53}{space 2} .0242192{col 64}{space 1}   -0.46{col 73}{space 3}0.646{col 81}{space 4}-.0586043{col 94}{space 3} .0363332
{txt}{space 32}lnitemp {c |}{col 41}{res}{space 2} .2358351{col 53}{space 2} .2641427{col 64}{space 1}    0.89{col 73}{space 3}0.372{col 81}{space 4} -.281875{col 94}{space 3} .7535452
{txt}{space 29}lnprojsize {c |}{col 41}{res}{space 2} .0683082{col 53}{space 2}  .033223{col 64}{space 1}    2.06{col 73}{space 3}0.040{col 81}{space 4} .0031924{col 94}{space 3}  .133424
{txt}{space 28}projects_no {c |}{col 41}{res}{space 2} .0014935{col 53}{space 2} .0019296{col 64}{space 1}    0.77{col 73}{space 3}0.439{col 81}{space 4}-.0022885{col 94}{space 3} .0052755
{txt}{space 32}accrate {c |}{col 41}{res}{space 2} .5839865{col 53}{space 2} .5703192{col 64}{space 1}    1.02{col 73}{space 3}0.306{col 81}{space 4}-.5338186{col 94}{space 3} 1.701792
{txt}{space 24}newprojects_pct {c |}{col 41}{res}{space 2}-.8343548{col 53}{space 2} .1761212{col 64}{space 1}   -4.74{col 73}{space 3}0.000{col 81}{space 4}-1.179546{col 94}{space 3}-.4891635
{txt}{space 37}y1 {c |}{col 41}{res}{space 2}-1.172977{col 53}{space 2}  .108297{col 64}{space 1}  -10.83{col 73}{space 3}0.000{col 81}{space 4}-1.385235{col 94}{space 3}-.9607186
{txt}{space 37}y2 {c |}{col 41}{res}{space 2} -1.16547{col 53}{space 2} .1288102{col 64}{space 1}   -9.05{col 73}{space 3}0.000{col 81}{space 4}-1.417934{col 94}{space 3} -.913007
{txt}{space 37}y3 {c |}{col 41}{res}{space 2}-1.100203{col 53}{space 2} .1197185{col 64}{space 1}   -9.19{col 73}{space 3}0.000{col 81}{space 4}-1.334847{col 94}{space 3}-.8655596
{txt}{space 37}y4 {c |}{col 41}{res}{space 2}-.8201698{col 53}{space 2} .1202935{col 64}{space 1}   -6.82{col 73}{space 3}0.000{col 81}{space 4}-1.055941{col 94}{space 3}-.5843989
{txt}{space 37}y5 {c |}{col 41}{res}{space 2}-.9603445{col 53}{space 2} .1112069{col 64}{space 1}   -8.64{col 73}{space 3}0.000{col 81}{space 4}-1.178306{col 94}{space 3} -.742383
{txt}{space 37}y6 {c |}{col 41}{res}{space 2}-.8970484{col 53}{space 2}  .104207{col 64}{space 1}   -8.61{col 73}{space 3}0.000{col 81}{space 4} -1.10129{col 94}{space 3}-.6928065
{txt}{space 37}y7 {c |}{col 41}{res}{space 2}        0{col 53}{txt}  (omitted)
{space 22}involuntary_gs13b {c |}{col 41}{res}{space 2} 43.78408{col 53}{space 2} 31.01012{col 64}{space 1}    1.41{col 73}{space 3}0.158{col 81}{space 4}-16.99464{col 94}{space 3} 104.5628
{txt}{space 20}involgs13_los15pctb {c |}{col 41}{res}{space 2} 31.51251{col 53}{space 2}  53.4985{col 64}{space 1}    0.59{col 73}{space 3}0.556{col 81}{space 4}-73.34261{col 94}{space 3} 136.3676
{txt}{space 17}activityno_permil_pctb {c |}{col 41}{res}{space 2} .8575665{col 53}{space 2} .3902828{col 64}{space 1}    2.20{col 73}{space 3}0.028{col 81}{space 4} .0926262{col 94}{space 3} 1.622507
{txt}{space 31}lnitempb {c |}{col 41}{res}{space 2}-.1925907{col 53}{space 2} .2685992{col 64}{space 1}   -0.72{col 73}{space 3}0.473{col 81}{space 4}-.7190355{col 94}{space 3}  .333854
{txt}{space 28}lnprojsizeb {c |}{col 41}{res}{space 2}-.0196736{col 53}{space 2} .0700301{col 64}{space 1}   -0.28{col 73}{space 3}0.779{col 81}{space 4}-.1569301{col 94}{space 3}  .117583
{txt}{space 27}projects_nob {c |}{col 41}{res}{space 2}-.0035584{col 53}{space 2} .0029376{col 64}{space 1}   -1.21{col 73}{space 3}0.226{col 81}{space 4} -.009316{col 94}{space 3} .0021993
{txt}{space 31}accrateb {c |}{col 41}{res}{space 2}-1.561293{col 53}{space 2} 1.287509{col 64}{space 1}   -1.21{col 73}{space 3}0.225{col 81}{space 4}-4.084765{col 94}{space 3} .9621782
{txt}{space 23}newprojects_pctb {c |}{col 41}{res}{space 2}-.8326972{col 53}{space 2}  .400722{col 64}{space 1}   -2.08{col 73}{space 3}0.038{col 81}{space 4}-1.618098{col 94}{space 3}-.0472966
{txt}{space 28}cio_tenureb {c |}{col 41}{res}{space 2}-.0121463{col 53}{space 2} .0590237{col 64}{space 1}   -0.21{col 73}{space 3}0.837{col 81}{space 4}-.1278307{col 94}{space 3}  .103538
{txt}{space 17}dmeratio_totalspending {c |}{col 41}{res}{space 2} .0987915{col 53}{space 2} .4595103{col 64}{space 1}    0.21{col 73}{space 3}0.830{col 81}{space 4}-.8018321{col 94}{space 3} .9994152
{txt}{space 16}dmeratio_totalspendingb {c |}{col 41}{res}{space 2} .2530247{col 53}{space 2} .5266441{col 64}{space 1}    0.48{col 73}{space 3}0.631{col 81}{space 4}-.7791787{col 94}{space 3} 1.285228
{txt}{space 34}_cons {c |}{col 41}{res}{space 2} .4857738{col 53}{space 2} .4545065{col 64}{space 1}    1.07{col 73}{space 3}0.285{col 81}{space 4}-.4050425{col 94}{space 3}  1.37659
{txt}{hline 40}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est6{txt} stored)

{com}. 
. * Managerial Involuntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Service Length
. margins , dydx(involuntary_gs13) at(involgs13_los15pct=($involgs13_los_p10 $involgs13_los_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.01}}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}   Contrast{col 30} Delta-method{col 41}    Una{col 50}djusted{col 58}          Una{col 71}djusted
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary_gs13 {txt}{c |}
{space 13}_at {c |}
{space 9}2 vs 1  {c |}{col 18}{res}{space 2}-5.736704{col 30}{space 2} 4.330398{col 41}{space 1}   -1.32{col 50}{space 3}0.185{col 58}{space 4}-14.22413{col 71}{space 3} 2.750721
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. * [FIGURE 4F]
. margins , dydx(involuntary_gs13) at(involgs13_los15pct=(0(0.01)0.05))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.01}}
{lalign 7:3._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.02}}
{lalign 7:4._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.03}}
{lalign 7:5._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.04}}
{lalign 7:6._at: }{space 0}{lalign 16:involgs13_los1~t} = {res:{ralign 3:.05}}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}{col 30} Delta-method
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary_gs13 {txt}{c |}
{space 13}_at {c |}
{space 14}1  {c |}{col 18}{res}{space 2}-.0397338{col 30}{space 2} 2.659426{col 41}{space 1}   -0.01{col 50}{space 3}0.988{col 58}{space 4}-5.252114{col 71}{space 3} 5.172646
{txt}{space 14}2  {c |}{col 18}{res}{space 2}-5.776437{col 30}{space 2} 5.082019{col 41}{space 1}   -1.14{col 50}{space 3}0.256{col 58}{space 4}-15.73701{col 71}{space 3} 4.184137
{txt}{space 14}3  {c |}{col 18}{res}{space 2}-12.16566{col 30}{space 2} 9.385254{col 41}{space 1}   -1.30{col 50}{space 3}0.195{col 58}{space 4}-30.56042{col 71}{space 3} 6.229095
{txt}{space 14}4  {c |}{col 18}{res}{space 2}-18.43982{col 30}{space 2} 12.16839{col 41}{space 1}   -1.52{col 50}{space 3}0.130{col 58}{space 4}-42.28943{col 71}{space 3} 5.409781
{txt}{space 14}5  {c |}{col 18}{res}{space 2}-23.76974{col 30}{space 2}  11.3904{col 41}{space 1}   -2.09{col 50}{space 3}0.037{col 58}{space 4}-46.09452{col 71}{space 3}-1.444964
{txt}{space 14}6  {c |}{col 18}{res}{space 2}-27.45144{col 30}{space 2} 7.189507{col 41}{space 1}   -3.82{col 50}{space 3}0.000{col 58}{space 4}-41.54262{col 71}{space 3}-13.36027
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.01 "1" 0.02 "2" 0.03 "3" 0.04 "4" 0.05 "5",format(%9.0f)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Departing Managers' Service Length (>15 years, %)", margin(t=1)  size(11pt)) ///
> title("{c -(}bf: Figure 4F. Conditional Turnover Effects by Departures' Service Length (H2){c )-}" "{c -(}bf:Managerial Involuntary Turnover Model [MODEL 22]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram involgs13_los15pct if involgs13_los15pct<=0.05, yaxis(2) xaxis(1)  fcolor(gray%20) yscale(axis(2) alt) lc(gray%10) xlabel(0 "0" 0.01 "1" 0.02 "2" 0.03 "3" 0.04 "4" 0.05 "5",format(%9.0f)) ylabel( ,format(%9.1f) angle(0) ) ylabel(,format(%9.0f) angle(0) axis(2)) ytitle(,axis(2))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:involgs13_los15pct}{p_end}
{res}{txt}
{com}. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4F.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 4F.04-29-2025.gph} saved

{com}. 
. 
. * [MODEL 23, FIGURE 5F] H3 : DOES NOT CONVERGE.
. 
. eststo: xtgee delay_pct c.involuntary_gs13##i.cio5 c.involgs13_los15pct  activityno_permil_pct cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.involuntary_gs13b  c.involgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb   if e(sample), fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:1.cio5#c.involuntary_gs13} != 0 predicts success perfectly;
      {bf:1.cio5#c.involuntary_gs13} omitted and 10 obs not used.

note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}8.9791582
{txt}Iteration 2:  Tolerance = {res}.55343416
{txt}Iteration 3:  Tolerance = {res}.50729392
{txt}Iteration 4:  Tolerance = {res}.67310628
{txt}Iteration 5:  Tolerance = {res}1.213259
{txt}Iteration 6:  Tolerance = {res}.4027022
{txt}Iteration 7:  Tolerance = {res}.16364876
{txt}Iteration 8:  Tolerance = {res}.08079983
{txt}Iteration 9:  Tolerance = {res}.04311218
{txt}Iteration 10: Tolerance = {res}.024007
{txt}Iteration 11: Tolerance = {res}.01373375
{txt}Iteration 12: Tolerance = {res}.00798871
{txt}Iteration 13: Tolerance = {res}.00469706
{txt}Iteration 14: Tolerance = {res}.00278225
{txt}Iteration 15: Tolerance = {res}.00165647
{txt}Iteration 16: Tolerance = {res}.00098965
{txt}Iteration 17: Tolerance = {res}.0005927
{txt}Iteration 18: Tolerance = {res}.00035557
{txt}Iteration 19: Tolerance = {res}.00021356
{txt}Iteration 20: Tolerance = {res}.00012838
{txt}Iteration 21: Tolerance = {res}.00007722
{txt}Iteration 22: Tolerance = {res}.00004647
{txt}Iteration 23: Tolerance = {res}.00002797
{txt}Iteration 24: Tolerance = {res}.00001684
{txt}Iteration 25: Tolerance = {res}.00001014
{txt}Iteration 26: Tolerance = {res}6.106e-06
{txt}Iteration 27: Tolerance = {res}3.678e-06
{txt}Iteration 28: Tolerance = {res}2.215e-06
{txt}Iteration 29: Tolerance = {res}1.334e-06
{txt}Iteration 30: Tolerance = {res}8.036e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:512}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.6}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:27})} = {res}{ralign 6:297.36}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 89:(Std. err. adjusted for clustering on {res:a_id})}
{hline 24}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 25}{c |}{col 37}  Semirobust
{col 1}              delay_pct{col 25}{c |} Coefficient{col 37}  std. err.{col 49}      z{col 57}   P>|z|{col 65}     [95% con{col 78}f. interval]
{hline 24}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 7}involuntary_gs13 {c |}{col 25}{res}{space 2}  3.45284{col 37}{space 2} 8.958613{col 48}{space 1}    0.39{col 57}{space 3}0.700{col 65}{space 4}-14.10572{col 78}{space 3}  21.0114
{txt}{space 17}1.cio5 {c |}{col 25}{res}{space 2}-.1822706{col 37}{space 2} .0890369{col 48}{space 1}   -2.05{col 57}{space 3}0.041{col 65}{space 4}-.3567798{col 78}{space 3}-.0077614
{txt}{space 23} {c |}
cio5#c.involuntary_gs13 {c |}
{space 21}1  {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 23} {c |}
{space 5}involgs13_los15pct {c |}{col 25}{res}{space 2}-7.106507{col 37}{space 2} 11.52999{col 48}{space 1}   -0.62{col 57}{space 3}0.538{col 65}{space 4}-29.70487{col 78}{space 3} 15.49185
{txt}{space 2}activityno_permil_pct {c |}{col 25}{res}{space 2} .2168677{col 37}{space 2} .1987886{col 48}{space 1}    1.09{col 57}{space 3}0.275{col 65}{space 4}-.1727509{col 78}{space 3} .6064862
{txt}{space 13}cio_tenure {c |}{col 25}{res}{space 2} -.010234{col 37}{space 2} .0251544{col 48}{space 1}   -0.41{col 57}{space 3}0.684{col 65}{space 4}-.0595357{col 78}{space 3} .0390678
{txt}{space 16}lnitemp {c |}{col 25}{res}{space 2} .2806823{col 37}{space 2} .2541662{col 48}{space 1}    1.10{col 57}{space 3}0.269{col 65}{space 4}-.2174744{col 78}{space 3} .7788389
{txt}{space 13}lnprojsize {c |}{col 25}{res}{space 2} .0696253{col 37}{space 2} .0371587{col 48}{space 1}    1.87{col 57}{space 3}0.061{col 65}{space 4}-.0032044{col 78}{space 3}  .142455
{txt}{space 12}projects_no {c |}{col 25}{res}{space 2} .0022435{col 37}{space 2} .0021803{col 48}{space 1}    1.03{col 57}{space 3}0.303{col 65}{space 4}-.0020297{col 78}{space 3} .0065168
{txt}{space 16}accrate {c |}{col 25}{res}{space 2} .6348905{col 37}{space 2} .5750256{col 48}{space 1}    1.10{col 57}{space 3}0.270{col 65}{space 4} -.492139{col 78}{space 3}  1.76192
{txt}{space 8}newprojects_pct {c |}{col 25}{res}{space 2}-.8487305{col 37}{space 2} .1724231{col 48}{space 1}   -4.92{col 57}{space 3}0.000{col 65}{space 4}-1.186674{col 78}{space 3}-.5107874
{txt}{space 21}y1 {c |}{col 25}{res}{space 2}-1.167219{col 37}{space 2} .1099883{col 48}{space 1}  -10.61{col 57}{space 3}0.000{col 65}{space 4}-1.382792{col 78}{space 3}-.9516458
{txt}{space 21}y2 {c |}{col 25}{res}{space 2}-1.157361{col 37}{space 2} .1287932{col 48}{space 1}   -8.99{col 57}{space 3}0.000{col 65}{space 4}-1.409791{col 78}{space 3} -.904931
{txt}{space 21}y3 {c |}{col 25}{res}{space 2}-1.074072{col 37}{space 2} .1182134{col 48}{space 1}   -9.09{col 57}{space 3}0.000{col 65}{space 4}-1.305766{col 78}{space 3}-.8423778
{txt}{space 21}y4 {c |}{col 25}{res}{space 2}-.7930312{col 37}{space 2}   .12226{col 48}{space 1}   -6.49{col 57}{space 3}0.000{col 65}{space 4}-1.032657{col 78}{space 3}-.5534059
{txt}{space 21}y5 {c |}{col 25}{res}{space 2}-.9298928{col 37}{space 2} .1111247{col 48}{space 1}   -8.37{col 57}{space 3}0.000{col 65}{space 4}-1.147693{col 78}{space 3}-.7120925
{txt}{space 21}y6 {c |}{col 25}{res}{space 2}-.8757759{col 37}{space 2} .1044156{col 48}{space 1}   -8.39{col 57}{space 3}0.000{col 65}{space 4}-1.080427{col 78}{space 3}-.6711251
{txt}{space 21}y7 {c |}{col 25}{res}{space 2}        0{col 37}{txt}  (omitted)
{space 6}involuntary_gs13b {c |}{col 25}{res}{space 2}  39.9682{col 37}{space 2} 31.48975{col 48}{space 1}    1.27{col 57}{space 3}0.204{col 65}{space 4}-21.75058{col 78}{space 3}  101.687
{txt}{space 4}involgs13_los15pctb {c |}{col 25}{res}{space 2} 40.55138{col 37}{space 2} 54.14404{col 48}{space 1}    0.75{col 57}{space 3}0.454{col 65}{space 4}-65.56898{col 78}{space 3} 146.6717
{txt}{space 1}activityno_permil_pctb {c |}{col 25}{res}{space 2} .8027762{col 37}{space 2} .3888769{col 48}{space 1}    2.06{col 57}{space 3}0.039{col 65}{space 4} .0405914{col 78}{space 3} 1.564961
{txt}{space 15}lnitempb {c |}{col 25}{res}{space 2}-.2265514{col 37}{space 2} .2561009{col 48}{space 1}   -0.88{col 57}{space 3}0.376{col 65}{space 4}-.7284999{col 78}{space 3} .2753971
{txt}{space 12}lnprojsizeb {c |}{col 25}{res}{space 2}-.0400188{col 37}{space 2} .0786407{col 48}{space 1}   -0.51{col 57}{space 3}0.611{col 65}{space 4}-.1941517{col 78}{space 3} .1141141
{txt}{space 11}projects_nob {c |}{col 25}{res}{space 2}-.0027209{col 37}{space 2}  .003782{col 48}{space 1}   -0.72{col 57}{space 3}0.472{col 65}{space 4}-.0101334{col 78}{space 3} .0046916
{txt}{space 15}accrateb {c |}{col 25}{res}{space 2}-1.432354{col 37}{space 2} 1.286681{col 48}{space 1}   -1.11{col 57}{space 3}0.266{col 65}{space 4}-3.954203{col 78}{space 3} 1.089494
{txt}{space 7}newprojects_pctb {c |}{col 25}{res}{space 2}-.7716691{col 37}{space 2} .3899074{col 48}{space 1}   -1.98{col 57}{space 3}0.048{col 65}{space 4}-1.535873{col 78}{space 3}-.0074646
{txt}{space 12}cio_tenureb {c |}{col 25}{res}{space 2}-.0162061{col 37}{space 2} .0586406{col 48}{space 1}   -0.28{col 57}{space 3}0.782{col 65}{space 4}-.1311396{col 78}{space 3} .0987275
{txt}{space 1}dmeratio_totalspending {c |}{col 25}{res}{space 2}-.0262459{col 37}{space 2} .4762822{col 48}{space 1}   -0.06{col 57}{space 3}0.956{col 65}{space 4}-.9597419{col 78}{space 3} .9072501
{txt}dmeratio_totalspendingb {c |}{col 25}{res}{space 2} .3293998{col 37}{space 2}  .536942{col 48}{space 1}    0.61{col 57}{space 3}0.540{col 65}{space 4}-.7229872{col 78}{space 3} 1.381787
{txt}{space 18}_cons {c |}{col 25}{res}{space 2} .4444694{col 37}{space 2} .4406728{col 48}{space 1}    1.01{col 57}{space 3}0.313{col 65}{space 4}-.4192335{col 78}{space 3} 1.308172
{txt}{hline 24}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est7{txt} stored)

{com}. 
. * [FIGURE 5F]
. margins , dydx(involuntary_gs13) over(cio5)
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:512}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2col:Over:}{res:cio5}{p_end}
{p2colreset}{...}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}{col 30} Delta-method
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary_gs13 {txt}{c |}
{space 12}cio5 {c |}
{space 14}0  {c |}{col 18}{res}{space 2} 1.059567{col 30}{space 2} 2.751403{col 41}{space 1}    0.39{col 50}{space 3}0.700{col 58}{space 4}-4.333083{col 71}{space 3} 6.452217
{txt}{space 14}1  {c |}{col 18}{res}{space 2} .9778642{col 30}{space 2} 2.538591{col 41}{space 1}    0.39{col 50}{space 3}0.700{col 58}{space 4}-3.997682{col 71}{space 3} 5.953411
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(scatter) plotopt(mcolor(white)) ciopt(color(black%0))  yline(0, lcolor(red) lpattern(shortdash)) ///
> text(-5 0 "Not Reported", placement(south)) ///
> text(-5 1 "Not Reported", placement(south)) ///
> xlabel (-1 " " 0 "No Centralization" 1 "Centralization" 2 " ", labsize(11pt) noticks) plot1opts(msize(medlarge) msymbol(square)) legend(off) ///
> ylabel( ,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Project Delays (%)", size(11pt) margin(r=1)) xtitle("" ) ///
> title("{c -(}bf: Figure 5F. Conditional Turnover Effects by Agency Centralization (H3){c )-}" "{c -(}bf:Managerial Involuntary Turnover Model [MODEL 23]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram cio5 if cio5>1, yaxis(2) discrete barwidth(0.4) fcolor(gray%20) yscale(r(0 1.5) axis(2) alt) lc(gray%10) ytitle(,axis(2)) xlabel(-1 " " 0 "No Centralization" 1"Centralization" 2 " ", noticks) ylabel(0 "0" 0.5 "0.5" 1 "1.0" 1.5 "1.5" , format(%9.1f) angle(0) axis(2)) ylabel(, format(%9.1f) angle(0) axis(1))) xsize(6)
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:cio5}{p_end}
{res}{txt}
{com}. 
. *
. *
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5F.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 5F.04-29-2025.gph} saved

{com}. 
. 
. * [MODEL 24, FIGURE 6F] H4 : Turnover X Task Standardization
. 
. eststo: xtgee delay_pct c.involuntary_gs13##c.activityno_permil_pct i.cio5 c.involgs13_los15pct   cio_tenure  lnitemp lnprojsize projects_no accrate  newprojects_pct y1-y7 c.involuntary_gs13b  c.involgs13_los15pctb activityno_permil_pctb lnitempb lnprojsizeb projects_nob accrateb newprojects_pctb cio_tenureb   dmeratio_totalspending dmeratio_totalspendingb  , fam(bin) link(probit) corr(uns) vce(robust)
{txt}note: {bf:y7} omitted because of collinearity.

Iteration 1:  Tolerance = {res}3.6758676
{txt}Iteration 2:  Tolerance = {res}1.1409293
{txt}Iteration 3:  Tolerance = {res}2.5812442
{txt}Iteration 4:  Tolerance = {res}1.8408057
{txt}Iteration 5:  Tolerance = {res}.66404797
{txt}Iteration 6:  Tolerance = {res}.22498973
{txt}Iteration 7:  Tolerance = {res}.1038855
{txt}Iteration 8:  Tolerance = {res}.05371155
{txt}Iteration 9:  Tolerance = {res}.02938147
{txt}Iteration 10: Tolerance = {res}.01661184
{txt}Iteration 11: Tolerance = {res}.00958949
{txt}Iteration 12: Tolerance = {res}.00561032
{txt}Iteration 13: Tolerance = {res}.00331099
{txt}Iteration 14: Tolerance = {res}.00196515
{txt}Iteration 15: Tolerance = {res}.00117073
{txt}Iteration 16: Tolerance = {res}.00069917
{txt}Iteration 17: Tolerance = {res}.00041823
{txt}Iteration 18: Tolerance = {res}.00025045
{txt}Iteration 19: Tolerance = {res}.00015009
{txt}Iteration 20: Tolerance = {res}.00008999
{txt}Iteration 21: Tolerance = {res}.00005397
{txt}Iteration 22: Tolerance = {res}.00003238
{txt}Iteration 23: Tolerance = {res}.00001943
{txt}Iteration 24: Tolerance = {res}.00001166
{txt}Iteration 25: Tolerance = {res}6.995e-06
{txt}Iteration 26: Tolerance = {res}4.198e-06
{txt}Iteration 27: Tolerance = {res}2.519e-06
{txt}Iteration 28: Tolerance = {res}1.512e-06
{txt}Iteration 29: Tolerance = {res}9.075e-07

{txt}GEE population-averaged model{col 54}{txt}{lalign 16:Number of obs} = {res}{ralign 6:522}
{txt}Group and time vars: {res}a_id fy{txt}{col 54}{txt}{lalign 16:Number of groups} = {res}{ralign 6:92}
{txt}Family: {lalign 8:Binomial}{col 54}{lalign 16:Obs per group:}
Link:   {lalign 8:Probit}{col 54}{ralign 16:min} = {res}{ralign 6:1}
{txt}Correlation: {res}unstructured{txt}{col 54}{ralign 16:avg} = {res}{ralign 6:5.7}
{txt}{col 54}{ralign 16:max} = {res}{ralign 6:7}
{txt}{col 54}{txt}{lalign 16:Wald chi2({res:28})} = {res}{ralign 6:310.23}
{txt}Scale parameter = {res}1{txt}{col 54}{txt}{lalign 16:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 108:(Std. err. adjusted for clustering on {res:a_id})}
{hline 43}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 44}{c |}{col 56}  Semirobust
{col 1}                                 delay_pct{col 44}{c |} Coefficient{col 56}  std. err.{col 68}      z{col 76}   P>|z|{col 84}     [95% con{col 97}f. interval]
{hline 43}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 26}involuntary_gs13 {c |}{col 44}{res}{space 2} 9.779794{col 56}{space 2} 18.12657{col 67}{space 1}    0.54{col 76}{space 3}0.590{col 84}{space 4}-25.74764{col 97}{space 3} 45.30723
{txt}{space 21}activityno_permil_pct {c |}{col 44}{res}{space 2} .2577046{col 56}{space 2} .1913033{col 67}{space 1}    1.35{col 76}{space 3}0.178{col 84}{space 4} -.117243{col 97}{space 3} .6326522
{txt}{space 42} {c |}
c.involuntary_gs13#c.activityno_permil_pct {c |}{col 44}{res}{space 2}-35.71321{col 56}{space 2}  73.4345{col 67}{space 1}   -0.49{col 76}{space 3}0.627{col 84}{space 4}-179.6422{col 97}{space 3} 108.2158
{txt}{space 42} {c |}
{space 36}1.cio5 {c |}{col 44}{res}{space 2}-.1616635{col 56}{space 2}  .084922{col 67}{space 1}   -1.90{col 76}{space 3}0.057{col 84}{space 4}-.3281075{col 97}{space 3} .0047805
{txt}{space 24}involgs13_los15pct {c |}{col 44}{res}{space 2}  1.50403{col 56}{space 2} 15.44362{col 67}{space 1}    0.10{col 76}{space 3}0.922{col 84}{space 4}-28.76491{col 97}{space 3} 31.77297
{txt}{space 32}cio_tenure {c |}{col 44}{res}{space 2} -.008512{col 56}{space 2} .0243305{col 67}{space 1}   -0.35{col 76}{space 3}0.726{col 84}{space 4}-.0561989{col 97}{space 3} .0391749
{txt}{space 35}lnitemp {c |}{col 44}{res}{space 2}  .253968{col 56}{space 2} .2690996{col 67}{space 1}    0.94{col 76}{space 3}0.345{col 84}{space 4}-.2734576{col 97}{space 3} .7813936
{txt}{space 32}lnprojsize {c |}{col 44}{res}{space 2} .0664692{col 56}{space 2} .0338236{col 67}{space 1}    1.97{col 76}{space 3}0.049{col 84}{space 4} .0001762{col 97}{space 3} .1327622
{txt}{space 31}projects_no {c |}{col 44}{res}{space 2} .0014513{col 56}{space 2} .0019313{col 67}{space 1}    0.75{col 76}{space 3}0.452{col 84}{space 4}-.0023341{col 97}{space 3} .0052366
{txt}{space 35}accrate {c |}{col 44}{res}{space 2} .5942798{col 56}{space 2} .5728333{col 67}{space 1}    1.04{col 76}{space 3}0.300{col 84}{space 4}-.5284528{col 97}{space 3} 1.717012
{txt}{space 27}newprojects_pct {c |}{col 44}{res}{space 2} -.846352{col 56}{space 2} .1741287{col 67}{space 1}   -4.86{col 76}{space 3}0.000{col 84}{space 4}-1.187638{col 97}{space 3} -.505066
{txt}{space 40}y1 {c |}{col 44}{res}{space 2}-1.162452{col 56}{space 2} .1064371{col 67}{space 1}  -10.92{col 76}{space 3}0.000{col 84}{space 4}-1.371065{col 97}{space 3}-.9538388
{txt}{space 40}y2 {c |}{col 44}{res}{space 2}-1.155659{col 56}{space 2} .1274562{col 67}{space 1}   -9.07{col 76}{space 3}0.000{col 84}{space 4}-1.405469{col 97}{space 3}-.9058494
{txt}{space 40}y3 {c |}{col 44}{res}{space 2}-1.094614{col 56}{space 2} .1181709{col 67}{space 1}   -9.26{col 76}{space 3}0.000{col 84}{space 4}-1.326225{col 97}{space 3}-.8630031
{txt}{space 40}y4 {c |}{col 44}{res}{space 2}-.8183223{col 56}{space 2} .1200401{col 67}{space 1}   -6.82{col 76}{space 3}0.000{col 84}{space 4}-1.053596{col 97}{space 3}-.5830481
{txt}{space 40}y5 {c |}{col 44}{res}{space 2}-.9543037{col 56}{space 2} .1113213{col 67}{space 1}   -8.57{col 76}{space 3}0.000{col 84}{space 4}-1.172489{col 97}{space 3} -.736118
{txt}{space 40}y6 {c |}{col 44}{res}{space 2}-.8977498{col 56}{space 2} .1046989{col 67}{space 1}   -8.57{col 76}{space 3}0.000{col 84}{space 4}-1.102956{col 97}{space 3}-.6925437
{txt}{space 40}y7 {c |}{col 44}{res}{space 2}        0{col 56}{txt}  (omitted)
{space 25}involuntary_gs13b {c |}{col 44}{res}{space 2} 46.78731{col 56}{space 2} 32.31387{col 67}{space 1}    1.45{col 76}{space 3}0.148{col 84}{space 4}-16.54671{col 97}{space 3} 110.1213
{txt}{space 23}involgs13_los15pctb {c |}{col 44}{res}{space 2} 34.85457{col 56}{space 2} 54.00835{col 67}{space 1}    0.65{col 76}{space 3}0.519{col 84}{space 4}-70.99984{col 97}{space 3}  140.709
{txt}{space 20}activityno_permil_pctb {c |}{col 44}{res}{space 2} .8989318{col 56}{space 2} .3971218{col 67}{space 1}    2.26{col 76}{space 3}0.024{col 84}{space 4} .1205874{col 97}{space 3} 1.677276
{txt}{space 34}lnitempb {c |}{col 44}{res}{space 2} -.204628{col 56}{space 2} .2717977{col 67}{space 1}   -0.75{col 76}{space 3}0.452{col 84}{space 4}-.7373418{col 97}{space 3} .3280858
{txt}{space 31}lnprojsizeb {c |}{col 44}{res}{space 2}-.0181667{col 56}{space 2} .0700942{col 67}{space 1}   -0.26{col 76}{space 3}0.795{col 84}{space 4}-.1555488{col 97}{space 3} .1192153
{txt}{space 30}projects_nob {c |}{col 44}{res}{space 2}-.0037696{col 56}{space 2} .0029291{col 67}{space 1}   -1.29{col 76}{space 3}0.198{col 84}{space 4}-.0095106{col 97}{space 3} .0019714
{txt}{space 34}accrateb {c |}{col 44}{res}{space 2}-1.593844{col 56}{space 2} 1.295601{col 67}{space 1}   -1.23{col 76}{space 3}0.219{col 84}{space 4}-4.133175{col 97}{space 3} .9454881
{txt}{space 26}newprojects_pctb {c |}{col 44}{res}{space 2}-.7832897{col 56}{space 2} .3942717{col 67}{space 1}   -1.99{col 76}{space 3}0.047{col 84}{space 4}-1.556048{col 97}{space 3}-.0105315
{txt}{space 31}cio_tenureb {c |}{col 44}{res}{space 2}-.0092256{col 56}{space 2} .0593141{col 67}{space 1}   -0.16{col 76}{space 3}0.876{col 84}{space 4}-.1254791{col 97}{space 3} .1070279
{txt}{space 20}dmeratio_totalspending {c |}{col 44}{res}{space 2} .0668676{col 56}{space 2}  .465563{col 67}{space 1}    0.14{col 76}{space 3}0.886{col 84}{space 4}-.8456191{col 97}{space 3} .9793542
{txt}{space 19}dmeratio_totalspendingb {c |}{col 44}{res}{space 2}   .30659{col 56}{space 2} .5266862{col 67}{space 1}    0.58{col 76}{space 3}0.560{col 84}{space 4}-.7256961{col 97}{space 3} 1.338876
{txt}{space 37}_cons {c |}{col 44}{res}{space 2} .4101939{col 56}{space 2} .4436516{col 67}{space 1}    0.92{col 76}{space 3}0.355{col 84}{space 4}-.4593472{col 97}{space 3} 1.279735
{txt}{hline 43}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
({res}est8{txt} stored)

{com}. 
. * Managerial Involuntary Turnover Rate: Marginal Effect Differentials Based on Interdecile Change in Task Standardization
. margins , dydx(involuntary_gs13) at(activityno_permil_pct=($stdz_p10 $stdz_p90)) pwcompare(effects)
{res}
{txt}Pairwise comparisons of average marginal effects

{col 1}Model VCE: {res:Semirobust}{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2colreset}{...}
{lalign 7:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:0}}
{lalign 7:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 3:.71}}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}   Contrast{col 30} Delta-method{col 41}    Una{col 50}djusted{col 58}          Una{col 71}djusted
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary_gs13 {txt}{c |}
{space 13}_at {c |}
{space 9}2 vs 1  {c |}{col 18}{res}{space 2}-7.802899{col 30}{space 2} 16.09577{col 41}{space 1}   -0.48{col 50}{space 3}0.628{col 58}{space 4}-39.35003{col 71}{space 3} 23.74423
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. 
. 
. * [FIGURE 6F]
. margins , dydx(involuntary_gs13) at(activityno_permil_pct=(0(0.1)1))
{res}
{txt}{col 1}Average marginal effects{col 60}{lalign 13:Number of obs}{col 73} = {res}{ralign 3:522}
{txt}{col 1}Model VCE: {res:Semirobust}

{txt}{p2colset 1 13 13 2}{...}
{p2col:Expression:}{res:Pr(delay_pct != 0), predict()}{p_end}
{p2col:dy/dx wrt:}{res:involuntary_gs13}{p_end}
{p2colreset}{...}
{lalign 8:1._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:0}}
{lalign 8:2._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.1}}
{lalign 8:3._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.2}}
{lalign 8:4._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.3}}
{lalign 8:5._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.4}}
{lalign 8:6._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.5}}
{lalign 8:7._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.6}}
{lalign 8:8._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.7}}
{lalign 8:9._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.8}}
{lalign 8:10._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:.9}}
{lalign 8:11._at: }{space 0}{lalign 16:activityno_per~t} = {res:{ralign 2:1}}

{res}{txt}{hline 17}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 18}{c |}{col 30} Delta-method
{col 18}{c |}      dy/dx{col 30}   std. err.{col 42}      z{col 50}   P>|z|{col 58}     [95% con{col 71}f. interval]
{hline 17}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}involuntary_gs13 {txt}{c |}
{space 13}_at {c |}
{space 14}1  {c |}{col 18}{res}{space 2} 2.850734{col 30}{space 2} 5.304903{col 41}{space 1}    0.54{col 50}{space 3}0.591{col 58}{space 4}-7.546685{col 71}{space 3} 13.24815
{txt}{space 14}2  {c |}{col 18}{res}{space 2} 1.834861{col 30}{space 2} 3.490754{col 41}{space 1}    0.53{col 50}{space 3}0.599{col 58}{space 4}-5.006892{col 71}{space 3} 8.676613
{txt}{space 14}3  {c |}{col 18}{res}{space 2}  .789787{col 30}{space 2} 2.264302{col 41}{space 1}    0.35{col 50}{space 3}0.727{col 58}{space 4}-3.648164{col 71}{space 3} 5.227738
{txt}{space 14}4  {c |}{col 18}{res}{space 2} -.283354{col 30}{space 2}  2.76509{col 41}{space 1}   -0.10{col 50}{space 3}0.918{col 58}{space 4}-5.702831{col 71}{space 3} 5.136123
{txt}{space 14}5  {c |}{col 18}{res}{space 2}-1.383415{col 30}{space 2}  4.51499{col 41}{space 1}   -0.31{col 50}{space 3}0.759{col 58}{space 4}-10.23263{col 71}{space 3} 7.465803
{txt}{space 14}6  {c |}{col 18}{res}{space 2}-2.509243{col 30}{space 2}  6.63043{col 41}{space 1}   -0.38{col 50}{space 3}0.705{col 58}{space 4}-15.50465{col 71}{space 3} 10.48616
{txt}{space 14}7  {c |}{col 18}{res}{space 2}-3.659677{col 30}{space 2} 8.886954{col 41}{space 1}   -0.41{col 50}{space 3}0.680{col 58}{space 4}-21.07779{col 71}{space 3} 13.75843
{txt}{space 14}8  {c |}{col 18}{res}{space 2}-4.833538{col 30}{space 2} 11.22583{col 41}{space 1}   -0.43{col 50}{space 3}0.667{col 58}{space 4}-26.83576{col 71}{space 3} 17.16869
{txt}{space 14}9  {c |}{col 18}{res}{space 2}-6.029614{col 30}{space 2} 13.62583{col 41}{space 1}   -0.44{col 50}{space 3}0.658{col 58}{space 4}-32.73574{col 71}{space 3} 20.67652
{txt}{space 13}10  {c |}{col 18}{res}{space 2}-7.246643{col 30}{space 2} 16.07696{col 41}{space 1}   -0.45{col 50}{space 3}0.652{col 58}{space 4}-38.75692{col 71}{space 3} 24.26363
{txt}{space 13}11  {c |}{col 18}{res}{space 2}-8.483298{col 30}{space 2} 18.57332{col 41}{space 1}   -0.46{col 50}{space 3}0.648{col 58}{space 4}-44.88634{col 71}{space 3} 27.91974
{txt}{hline 17}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}{txt}
{com}. marginsplot, recast(connected) plot1opts(lcolor(black%60) mcolor(black%60))  ciopt(color(black%30)) yline(0, lcolor(red) lpattern(shortdash)) recastci(rarea) ///
> legend(off) ylabel(,format(%9.1f) angle(0)) /// ///
> xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0)) ///
> ytitle("Marginal Change in Delayed Projects (%)", size(11pt) margin(r=1)) xtitle("Proportion of Standardized Projects (%)", size(11pt) margin(t=1)) ///
> title("{c -(}bf: Figure 6F. Conditional Turnover Effects by Task Standardization (H4){c )-}" "{c -(}bf:Managerial Involuntary Turnover Model [MODEL 24]{c )-}", size(12pt) margin(medium)) ///
> addplot(histogram activityno_permil_pct , yaxis(2) xaxis(1) fcolor(gray%20) lc(gray%10) yscale(axis(2) alt) xlabel(0 "0" 0.5 "50" 1 "100",format(%9.0f)) ylabel(,format(%9.1f) angle(0))  ylabel( ,format(%9.1f) angle(0) axis(2)) ytitle(,  axis(2))) xsize(6) 
{res}
{text}{p 0 0 2}Variables that uniquely identify margins: {bf:activityno_permil_pct}{p_end}
{res}{txt}
{com}. 
. graph save "Graph" "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6F.04-29-2025.gph", replace
{res}{txt}file {bf:C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\GRAPHICS\FIGURE 6F.04-29-2025.gph} saved

{com}. 
. *
. *
. ****************************************************************************************************************************************************
. *
. cd "C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES"
{res}C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\TABLES
{txt}
{com}. 
. esttab est1 est2 est3 est4 using ManuscriptTable3.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 3. Fractional Panel Probit Analysis of the Impact of Managerial Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 17" "Model 18" "Model 19" "Model 20" )  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) replace
{res}{txt}(output written to {browse  `"ManuscriptTable3.04-29-2025.rtf"'})

{com}. 
. 
. esttab est5 est6 est7 est8 using ManuscriptTable3.04-29-2025.rtf, ///
> cells(b(star fmt(%9.3f)) se(par) p(fmt(3) par("[" "]"))) starl( * 0.10 ** 0.05 *** 0.010) label ///
> title(TABLE 3. Fractional Panel Probit Analysis of the Impact of Managerial Voluntary and Involuntary Turnover on IT Project Delays in U.S. Federal Agencies, FY2015−FY2021) ///
> nonumbers mtitles("Model 21" "Model 22" "Model 23" "Model 24")  ///
> stats(year qic pseudor chi2 N, fmt(2) label("Year Fixed Effects" "QIC" "Pseudo R-Squared" "Wald" "N")) append
{res}{txt}(output written to {browse  `"ManuscriptTable3.04-29-2025.rtf"'})

{com}. ***************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
. log close
      {txt}name:  {res}<unnamed>
       {txt}log:  {res}C:\Users\hongj\Dropbox\2024 SPRING PROJECTS\Organizational Memory Paper\JPART Replication Datafiles\OUTPUT\OUTPUT.Organizational Memory.MANUSCRIPT.04-30-2025.smcl
  {txt}log type:  {res}smcl
 {txt}closed on:  {res}30 Apr 2025, 19:26:16
{txt}{.-}
{smcl}
{txt}{sf}{ul off}